Divergent biomass partitioning to aboveground and belowground across forests in China

Divergent biomass partitioning to aboveground and belowground across forests in China Abstract Aims Belowground to aboveground biomass (BGB/AGB) ratio is a highly valued parameter of the terrestrial carbon cycle and productivity. However, it remains far from clear whether plant biomass partitioning to aboveground and belowground is isometric (equal partitioning) or allometric (unequal partitioning) at community levels and what factors are necessary in order to regulate the partitioning. This study aimed to comprehensively find out the patterns of biomass partitioning and their regulatory factors across forests in China. Methods The data of AGB and BGB were compiled from 1542 samples for communities across forests in China. Standardized major axis regression was conducted to examine whether AGB and BGB were allocated isometrically or allometrically at a community level. Redundancy analysis was used to analyze the relationships of BGB/AGB ratio with climatic factors and soil properties. Important Findings We found that the slopes of the relationship between logAGB and logBGB were not always comparable to 1.0 (isometric allocation) at community levels, including primary forest, secondary forest, and planted forest. Meanwhile, samples in clay, loam, and sand soil types also presented the same phenomenon. Furthermore, the radically different allocations of AGB and BGB were found in northern and southern China. Environmental factors totally explained 3.86% of the variations in the BGB/AGB ratio at the community level, which include the mean annual precipitation, mean annual temperature, potential water deficit index, soil carbon content, soil nitrogen content, soil clay, soil loam, soil sand, soil pH, and soil bulk density. In addition, the environmental factors also have effects on the BGB/AGB ratio in other categories. The patterns revealed in this study are helpful for better understanding biomass partitioning and spreading the carbon circle models. biomass partitioning, isometric partitioning, BGB/AGB ratio, environmental factors, climatic zone, Chinese forest INTRODUCTION The belowground to aboveground biomass (BGB/AGB) ratio is a variable reflecting a plant’s response and adaption strategies to environmental stress and is also an important parameter for the terrestrial carbon cycle (Jackson et al. 2000; Wolf et al. 2011). Due to limited study of belowground part with great difficulty and high cost, biomass partitioning has not been well constrained in models, which induces large uncertainties in simulating the global carbon cycle and its responses to climate change (Wolf et al. 2011). It remains contentious whether biomass partitioning to belowground and aboveground is isometric or allometric, although it has been extensively examined at community levels in previous studies (Enquist 2002; McCarthy et al. 2007; McConnaughay et al. 1999; Shipley et al. 2002). Even less clear is how much of an impact can environmental factors have. Biomass partitioning is usually quantified as a slope of log-transformed aboveground biomass (logAGB) versus log-transformed belowground biomass (logBGB), which can be statistically identical to either 1 (isometric biomass partitioning) or not (allometric biomass partitioning). Regarding the isometric biomass partitioning, Cheng et al. (2007) reported that the relationships between logAGB and logBGB follow an isometric relationship across forested communities in China. Further, BGB/AGB ratio also did not have any significant linear trend with stand age across various forest ecosystems (Yang et al. 2011). Meanwhile, the isometric biomass partitioning has been verified in other ecosystems, for instance, alpine grasslands (Yang et al. 2009) and all grasslands in China (Yang et al. 2010). Nevertheless, Wang et al. (2008) found that the BGB/AGB ratio had negative relationships with water availability, stand age, height and volume, and hold, which had significant effects on the climate and ontogeny of biomass allocation. In addition, soil nutrients play important roles in biomass allocation; the lower phosphorus availability results in the higher allocation of belowground biomass and production (Castaneda-Moya et al. 2013), and light availability also affects plant biomass allocation (Larcher et al. 2012). These partitioning patterns can be seen as a result of increased allocation by plants of photosynthates to roots in low resource environments. This may also give rise to an increase in water and nutrient absorption for survival. (Lobos-Catalan et al. 2014; Sun et al. 2016). Though there is extensive knowledge on the partitioning patterns of AGB and BGB there is still need for integrated research at community level that investigates the influence of abiotic factors on biomass partitioning. To our knowledge, forests contribute largely to terrestrial ecosystem carbon uptake and thus are crucial components in regulating global and regional carbon cycles (Pan et al. 2011; Piao et al. 2005; Zhao et al. 2012). A wide range of forests can be found in China, from conifer to tropical forests along with other forest types (Fang et al. 2012) with the high mountain forests in the Tibetan Plateau exceptionally diverse when compared to other countries in the world. This diverse and varied range of forests with large spatial variations in biomass, climate and soil texture along with different factors for biomass partitioning offer natural experimental sites for research on structure and function of forest ecosystems. Forests in China extensively represent all forest types in the world from conifer forest to tropical forest as well as other types (Fang et al. 2012), whose diversity in China is even more than that in other countries of the world for high mountain forests in the Tibetan Plateau of China. Thus, China owns natural experimental fields for studying the structure and function of forest ecosystems and the controlling factors for biomass partitioning at the regional scale for the large spatial variations of biomass, climate and soil texture across all types of forests (Luo et al. 2012; Ren et al. 2011; Zhou et al. 2002). In this study, we aimed to examine variations in biomass partitioning across different forest ecosystems in China and their impact factors. The specific objectives were: (i) to investigate whether AGB scales isometrically or allometrically with BGB at a community level; (ii) to compare biomass partitioning patterns among different climatic zones; and (iii) to comprehensively examine the impacts of climate factors and soil properties, on biomass partitioning. MATERIAL AND METHODS Data compilation Biomass data for this study also utilized data that was collected from previous studies (Wang et al. 2014), with the total number of samples accounting to 1542 in the final database from the range of forests in China (see online supplementary Table S1). The following criteria were used to screen and filter the documents: (i) inclusion of both coarse and fine roots in BGB; (ii) all the samples were collected from mature forests; and (iii) the community biomass was calculated using the ‘standard tree’ method (Hui et al. 2014). Our database consisted of geographic location (longitude and latitude) and two target variables (AGB and BGB). Climatic variables [i.e. annual mean temperature, annual mean precipitation, potential water deficit index] and soil properties [i.e. soil nitrogen density (N), pH, soil carbon density (C), soil bulk density (bulk) and soil texture] were collected to analyze the relationships of environmental factors with AGB/BGB and their ratio in each study paper. The soil variables were collected from the data set of Zhu (2006) and Post (2000). Study area The study sites include all the forest types in China, which span a wide range of environmental conditions (Gao et al. 2013). The study area covered from 80.24° to 130.89°E in longitude and ranged from 41.28° to 44.443°N in latitude (Fig. 1). According to the category of physical geography in China (Physical Geography in China Editorial Board of Chinese Academy of Sciences 1985), climatic zones in China include the northern and southern zones. Meanwhile, the B line is 0°C isothermal line in January, and 800 mm precipitation line (Fig. 1). Figure 1: View largeDownload slide spatial distribution of sampling sites. A total of 1543 samples of aboveground biomass (AGB) and belowground biomass (BGB) were compiled from the wide ranging forest types in China. The red solid circles, yellow solid circles and green solid circles represent primary forest, planted forest and secondary forest, respectively. Some plots are not visible as there are instances of plot overlap. Line A is the boundary between the monsoon and non-monsoon and 400 mm of precipitation line. Line B is 0°C isothermal line in January and 800 mm precipitation line. Line C is the precipitation and topography line (Kunlun Mountains, Qilian Mountains, Hengduan Mountains). Figure 1: View largeDownload slide spatial distribution of sampling sites. A total of 1543 samples of aboveground biomass (AGB) and belowground biomass (BGB) were compiled from the wide ranging forest types in China. The red solid circles, yellow solid circles and green solid circles represent primary forest, planted forest and secondary forest, respectively. Some plots are not visible as there are instances of plot overlap. Line A is the boundary between the monsoon and non-monsoon and 400 mm of precipitation line. Line B is 0°C isothermal line in January and 800 mm precipitation line. Line C is the precipitation and topography line (Kunlun Mountains, Qilian Mountains, Hengduan Mountains). Data analysis To examine the allocation pattern (isometric or allometric) of AGB and BGB across different field sites and climatic zones at community levels the standardized major axis (SRMA) regression was conducted (Enquist et al. 2002). The regression relationship of AGB = βBGBα or LogAGB = αLogBGB + Logβ was used to describe the relationship between AGB and BGB, where α is the slope and Logβ is the y-intercept (Li et al. 2006). The Standardized Major Axis Tests & Routines Version 2.0 software package was used to determine the scaling slope and y-intercept of the function (Warton et al. 2006). The relationship between AGB and BGB was considered isometric if a 95% confidence interval (CI) of the scaling slope covered 1.0 and vice versa. Relationship among multiple interacting variables can be effectively evaluated using redundancy analysis (RA; Chen et al. 2016; Vandenwollenberg 1977); thus, the RA was used to analyze the relationships of environmental factors with BGB/AGB, and environmental factors were partitioned into two explanatory variable groups (climatic factors and soil factors). This process was operated calling program ‘vegan’ in the R software package (R Development Core Team, 2016). RESULTS Variations in BGB/AGB ratio At the community level the BGB/AGB ratio showed a mean value of 0.24 and variations ranging from 0.02 to 1.96 across all forest types (Fig. 2). The mean values of BGB/AGB ratio in clay, loam and sand soil types were 0.23, 0.23 and 0.26, respectively, while that in planted, primary and secondary forest BGB were 0.24, 0.25 and 0.27 separately. The maximum (ranging from 0.03 to 1.96) and minimum (ranging from 0.05 to 0.50) of variation were found in sand and clay soil types. Figure 2: View largeDownload slide the frequency distribution of belowground biomass (BGB)/aboveground biomass (AGB) in clay, loam, and sand soil types and in planted forest, primary forest, secondary forest and all samples in China at the community level. Figure 2: View largeDownload slide the frequency distribution of belowground biomass (BGB)/aboveground biomass (AGB) in clay, loam, and sand soil types and in planted forest, primary forest, secondary forest and all samples in China at the community level. Biomass partitioning patterns in different forest types BGB and AGB had a relationship being characterized by the log-transformed linear function of LogAGB = 0.71 + 0.96LogBGB (R2 = 0.84, P < 0.01) in planted forest. The fitting model had a slope of 0.96 at 95% CI between 0.93 and 0.98 (Fig. 3A), which followed allometric allocation, and slope was significantly different from 1. In primary forest, the relationship between BGB and AGB was characterized by the log-transformed linear function of LogAGB = 0.65 + 0.98LogBGB (R2 = 0.83, P > 0.05). The slope of the fitting model was 0.98, with a 95% CI of 0.94–1.02 (Fig. 3B), which followed isometric allocation. BGB showed a linear relationship with AGB in secondary forests after it was log transformed and characterized in function of LogAGB = 0.69 + 0.93LogBGB (R2 = 0.84, P > 0.05). The slope of the fitting model was 0.83, with a 95% CI of 0.84–1.02 (Fig. 3C), which followed isometric allocation. The log-transformed linear function of LogAGB = 0.69 + 0.96LogBGB was used to characterize the relationship between BGB and RGB, across different forest types. The fitting model had a slope of 0.96, with a 95% CI of 0.94–0.98 (Fig. 3D) that indicated a allometric allocation pattern (P < 0.001). Figure 3: View largeDownload slide in the planted forests the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level (A), primary forest (B), secondary forest (C) and all samples in China (D). The standardized major axis regression was used for all panels in the graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 3: View largeDownload slide in the planted forests the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level (A), primary forest (B), secondary forest (C) and all samples in China (D). The standardized major axis regression was used for all panels in the graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Biomass partitioning patterns in different soil types In clay types, the relationship between BGB and AGB was characterized by the log-transformed linear function of LogAGB = 0.63 + 1.03LogBGB (R2 = 0.90, P > 0.05). The slope of the fitting model was 1.03, with a 95% CI of 0.99–1.06 (Fig. 4). In loam soil types, the slope of the relationship between log AGB and log BGB was 1.02 with 95% CIs of 0.99–1.05. Thus, the biomass partitioning patterns in clay and loam soil types followed isometric allocation. Forests found in sandy soils showed opposite results with the relationship between BGB and AGB characterized by the log-transformed linear function of LogAGB = 0.62 + 1.02LogBGB (R2 = 0.79, P < 0.001), whose slope in the fitting model was 1.10, with a 95% CI of 1.04–1.17. Figure 4: View largeDownload slide under different soil types like clay, loan and sand the relationship between aboveground biomass (AGB) and belowground biomass (BGB). The standardized major axis regression was used for the graph. AGB and BGB was log 10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 4: View largeDownload slide under different soil types like clay, loan and sand the relationship between aboveground biomass (AGB) and belowground biomass (BGB). The standardized major axis regression was used for the graph. AGB and BGB was log 10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Biomass partitioning patterns in different climatic zones The relationships between BGB and AGB in different climate zones (Fig. 5). Slopes and the 95% CIs of fitting models were 0.99 (0.96–1.02) in northern China (Fig. 5A) and 1.05 (1.02–1.07) in southern China, respectively. Equations’ slopes had non-significant difference with 1 in northern China (Fig. 5B; P > 0.05), while significant difference the slope from 1 in southern China (P< 0.001), respectively. Figure 5: View largeDownload slide under different climate zones in China the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level. (A and B) Forest in northern China and southern China. Standardized major axis regression is used for all panels in graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 5: View largeDownload slide under different climate zones in China the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level. (A and B) Forest in northern China and southern China. Standardized major axis regression is used for all panels in graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Responses of BGB/AGB ratio to environmental factors The RA was used to further estimate the causal relationship among the interactive variables (Fig. 6). Result showed that all the environmental variables explained 3.86% of the variation in the BGB/AGB ratio of all forests in China (Fig. 6G). The effects of these environmental factors on the BGB/AGB ratio were as follows: climate factors (1.30%); soil factors (2.77%) and coupling effect (−0.21%). Meanwhile, the environmental factors have a small effect on the BGB/AGB ratio in other categories (Fig. 6A–B, D–F, and H–I). Figure 6: View largeDownload slide the contributions of environmental factors to BGB/AGB via redundancy analysis is presented in Venn diagrams. Contributions of climate factors and soil factors to the BGB/AGB in planted forest (A), primary forest (B), secondary forest (C), clay forest (D), loam forest (E), sand forest (F), forest in China (G), forest in northern China (H) and forest in southern China (I). Figure 6: View largeDownload slide the contributions of environmental factors to BGB/AGB via redundancy analysis is presented in Venn diagrams. Contributions of climate factors and soil factors to the BGB/AGB in planted forest (A), primary forest (B), secondary forest (C), clay forest (D), loam forest (E), sand forest (F), forest in China (G), forest in northern China (H) and forest in southern China (I). Mechanisms of observed differences in the root–shoot relationship Based on the above analysis, we developed a framework to explain observed differences of the root–shoot relationship (Fig. 7). Biomass allocation of plants could be influenced by environment stress. The scaling slope should be equal to 1 (isometric allocation) under non-stressful conditions. However, the scaling slope of plant can be flexible, which may vary from the minimum to the maximum and is regulated by environmental factors. The response of biomass allocation sometimes might produce ‘redundancy’ theoretically once it falls into ‘tolerance’ area. Plants were able to adjust the ratio of root–shoot within a certain range to adapt to the variation of environments. Figure 7: View largeDownload slide the mechanisms suggested explaining observed differences in the variation of BGB/AGB. Figure 7: View largeDownload slide the mechanisms suggested explaining observed differences in the variation of BGB/AGB. DISCUSSION Divergent biomass partitioning across forests in China Theoretically, optimal partitioning theory hold a view that plants prefer to partition more biomass to the organ to acquire the most limiting resource (Bloom et al. 1985; Thornley 1972). Contrary to this belief, the allometric biomass partitioning theory predicts that to minimize hydrodynamic resistance and transport duration and to maximize photosynthetic harvesting capacity and resource transport, plant may redistribute biomass (Enquist and Niklas 2002; McCarthy and Enquist 2007). However, in reality, both allometric (Roa-Fuentes et al. 2012) and isometric (Cheng and Niklas 2007; Poorter et al. 2000; Yang and Luo 2011) partitioning were documented in experimental studies at the community level. In this study, there is divergent biomass partitioning to aboveground and belowground across forests in China. At the community level, the co-existing species may have trade-off effects with each other to maximize the resource use efficiency (Kneitel et al. 2004), and the partitioning varies among different climate zones at community levels. We found that the northern climate zone (Fig. 5A) followed isometric pattern (P > 0.05), but the southern climate zone (Fig. 5B) was allometric allocation (slope = 1.05). In addition, biomass allocation in sand soils was statistically allometric but was isometric in clay and loam soils (Fig. 4). In sandy soil, high silt content will decrease macropores and increase bulk density (Archer et al. 2002), with a consequent limitation in root growth (Dostal et al. 2005). Our findings documented that different climate zones and soil types have different biomass partitioning patterns and indicate that climate elements and soil texture have impacts on biomass partitioning in forests. Furthermore, the phenomenon also was proved by RA (Fig. 6); 3.86% of the variation in BGB/AGB ratio was explained by environmental factors; and climate factors, soil factors and coupling effect were 1.30%, 2.77% and −0.21%, respectively. Meanwhile, the environmental factors explaining the variation in BGB/AGB ratio were 4.06%, 8.24%, 37.16%, 9.59%, 5.39%, 11.89%, 8.43% and 6.32% in planted forest, primary forest, secondary forest, clay forest, loam forest, sand forest, forest in northern China and forest in southern China. At large spatial and temporal scales precipitation and temperature act as limiting factors for the growth and distribution of vegetation (Sun et al. 2013b). Our result is consistent with previous studies that found temperature (Reich et al. 2014) and precipitation (Roa-Fuentes et al. 2012) was a driver for forest biomass distribution in aboveground and belowground biomass. Soil characteristics may be important factors influencing BGB/AGB besides climate factors (Becknell et al. 2014). Plants will allocate more biomass into roots for more uptake of soil nutrient (Gleason et al. 2011) or available water (Sun et al. 2013b), some local studies have shown changing allocation of BGB/AGB, regulated by soil characteristics [e.g. soil carbon (Sun and Wang 2016) and nitrogen (Makita et al. 2011; Sun et al. 2013a)]. Meanwhile, soil texture was significantly correlated with vegetation biomass in forest ecosystems, and the same result also was found in alpine grassland (Qin et al. 2015). In our study, our data also indicated BGB/AGB ratio was affected by soil factors (Fig. 6). This suggests that environmental factors have an impact on biomass partitioning in different categories across Chinese forests. Generally, this study more comprehensively revealed forest biomass partitioning in different forest types, soil types, and climate zones. The findings indicated that our results do not agree with the isometric allocation hypothesis regarding the relationship between AGB and BGB across Chinese forests. The BGB/AGB varied with the change of ambient conditions, and the biomass allocation also was governed by self-regulation (Wang et al. 2008; Zhang et al. 2016). Plants may have self-regulation capabilities within a certain range of environmental changes or disturbances in terms of biomass partitioning (Read et al. 2006). Once this self-regulation exceeds certain ‘minimum’ or ‘maximum’ limits, biomass would be over-partitioned to some extent (Kays et al. 1974). Hence, trade-off between biomass partitioning between aboveground and belowground is driven by both external ‘environmental filtering’ and internal adaptation strategies (Fig. 7). Implications for ecosystem modeling Assuming that a fixed fraction of the carbon assimilation is allocated to each organ makes it convenient to model carbon allocation (Ostle et al. 2009). Thus, to estimate future carbon budgets, dynamic global vegetation models have been used with fixed or only water dependent carbon allocation according to pipe theory, which was proposed based on the relationship between leaf water use and stem transport capacity (Ostle et al. 2009; Shinozaki et al. 1964). This leads to the large uncertainty in predicting the future global forest carbon balance (Purves et al. 2008). The simplification of biomass partitioning in models is due to a lack of empirical evidence on biomass partitioning and the difficulty in interpreting the available information under a wide range of conditions (Franklin et al. 2012). Meanwhile, most research about the carbon circle mostly based on model estimates, while the experimental data are penurious at large scale. With variations in climate and soil conditions, the different biomass partitioning patterns with climate zones and environmental factors were found. The results provide both empirical evidence and the environmental drivers for the biomass allocation in forests. The pipe model relationship is mainly governed by the internal relationship between plant organs for water; it is thus less invariant across environment and can’t be used to simulate effects of resource availability and climate change on C allocation in trees (Franklin et al. 2012). The results obtained show the implications on the development of allocation models. First, data to numerically parameterize allocation models are provided by allocation patters in different climate zones. Second, the controlling factors and their relationship with the biomass partitioning provide windows that are to be described mathematically in allocation models. Third, patterns revealed in the study provide a benchmark with which to evaluate and benchmark those quantitative predictions. Limitations of the current study Biomass partitioning is both determined and driven by environmental conditions and ontogeny (Schall et al. 2012, Wang et al. 2008). In the present study, we analyzed the relationships of environmental factors with BGB/AGB; as a matter of fact, the plants were able to self-regulate the ratio of BGB/AGB (roots, stem, and leaves) to adapt to an environment. In addition, a conceptual model (Fig. 7) may be a useful pathway to illustrate biomass allocation of plants to aboveground and belowground, and the allocated process is achieved through signal transduction to morphological or physiological response. The theory proposes that the sum of all modular responses of a plant to its local conditions along with the interaction effects due to integration gives the response of a plant to its environment. According to de Kroon et al. 2005 there may be rules that are seen as evolving traits targeted by natural selection—local response rules to environmental variation and the modular interaction rules. Consequently, the mechanism of self-regulation needs to be further explored in the future. CONCLUSIONS Synthesizing biomass partitioning patterns across forests in China, this study demonstrated that there were no general rules on biomass partitioning in the different soil types, forest types, and climate zones. Environmental factors (climate factors and soil factors) have effects on biomass partitioning. The results suggest that biomass partitioning is governed by external environmental factors and internal plant self-regulation. Our work provides useful constraints on biomass partitioning parameters, which may reduce the uncertainties in predicting the global carbon cycle. SUPPLEMENTARY MATERIAL Supplementary data are available at Journal of Plant Ecology online. FUNDING This study was jointly supported by CAS Strategic Priority Research Program (grant no. XDA05050702), the National Natural Science Foundation of China (31420103917, 31290221, 41661144045), the Thousand Youth Talents Plan Project, and the Open Fund of Key Laboratory of Ecosystem Network Observation and Modelling. ACKNOWLEDGEMENTS We thank Dr Dafeng Hui for his helpful comments that had improved this manuscript greatly. Meanwhile, we appreciate two anonymous reviewers for their professional and useful suggestions regarding our previous manuscript. Conflict of interest statement. None declared. REFERENCES Archer NAL Quinton JN Hess TM( 2002) Below-ground relationships of soil texture, roots and hydraulic conductivity in two-phase mosaic vegetation in south-east Spain. J Arid Environ  52: 535– 53. Google Scholar CrossRef Search ADS   Becknell JM Powers JS( 2014) Stand age and soils as drivers of plant functional traits and aboveground biomass in secondary tropical dry forest. Can J For Res  44: 604– 13. Google Scholar CrossRef Search ADS   Bloom AJ Chapin FS Mooney HA( 1985) Resource limitation in plants—an economic analogy. Annu Rev Ecol Syst  16: 363– 92. Google Scholar CrossRef Search ADS   Castaneda-Moya E Twilley RR Rivera-Monroy VH( 2013) Allocation of biomass and net primary productivity of mangrove forests along environmental gradients in the Florida coastal everglades, USA. For Ecol Manag  307: 226– 41. Google Scholar CrossRef Search ADS   Chen YL Ding JZ Peng YFet al.  ( 2016) Patterns and drivers of soil microbial communities in tibetan alpine and global terrestrial ecosystems. J Biogeogr  43: 2027– 39. Google Scholar CrossRef Search ADS   Cheng DL Niklas KJ( 2007) Above- and below-ground biomass relationships across 1534 forested communities. Ann Bot  99: 95– 102. Google Scholar CrossRef Search ADS PubMed  de Kroon H Huber H Stuefer JFet al.  ( 2005) A modular concept of phenotypic plasticity in plants. New Phytol  166: 73– 82. Google Scholar CrossRef Search ADS PubMed  Dostal P Breznova M Kozlickova Vet al.  ( 2005) Ant-induced soil modification and its effect on plant below-ground biomass. Pedobiologia  49: 127– 37. Google Scholar CrossRef Search ADS   Enquist BJ( 2002) Universal scaling in tree and vascular plant allometry: toward a general quantitative theory linking plant form and function from cells to ecosystems. Tree Physiol  22: 1045– 64. Google Scholar CrossRef Search ADS PubMed  Enquist BJ Niklas KJ( 2002) Global allocation rules for patterns of biomass partitioning in seed plants. Science  295: 1517– 20. Google Scholar CrossRef Search ADS PubMed  Fang JY Wang XP Liu YNet al.  ( 2012) Multi-scale patterns of forest structure and species composition in relation to climate in northeast china. Ecography  35: 1072– 82. Google Scholar CrossRef Search ADS   Franklin PA Bartels B( 2012) Restoring connectivity for migratory native fish in a New Zealand stream: effectiveness of retrofitting a pipe culvert. Aquat Conserv  22: 489– 97. Google Scholar CrossRef Search ADS   Gao YH Liu XX Min CCet al.  ( 2013) Estimation of the northsouth transect of eastern China forest biomass using remote sensing and forest inventory data. Int J Remote Sens  34: 5598– 610. Google Scholar CrossRef Search ADS   Gleason SM Read J Ares A( 2011) Biomass allocation and phosphorus economics of rain-forest seedlings: effects of fertilization and radiation on soil specialists and soil generalists. J Trop Ecol  27: 147– 61. Google Scholar CrossRef Search ADS   Hui DF Wang J Shen WJet al.  ( 2014) Near isometric biomass partitioning in forest ecosystems of china. PLOS ONE  9:e86550. Jackson RB Schenk HJ Jobbagy EGet al.  ( 2000) Belowground consequences of vegetation change and their treatment in models. Ecol Appl  10: 470– 83. Google Scholar CrossRef Search ADS   Kays S Harper JL( 1974) Regulation of plant and tiller density in a grass sward. J Ecol  62: 97– 105. Google Scholar CrossRef Search ADS   Kneitel JM Chase JM( 2004) Trade-offs in community ecology: linking spatial scales and species coexistence. Ecol Lett  7: 69– 80. Google Scholar CrossRef Search ADS   Larcher L Boeger MRT Marques MCM( 2012) Biomass allocation and shade tolerance in tree species of the Atlantic forest. Botany  90: 830– 8. Google Scholar CrossRef Search ADS   Li HT Han XG Wu JG( 2006) Variant scaling relationship for mass-density across tree-dominated communities. J Integr Plant Biol  48: 268– 77. Google Scholar CrossRef Search ADS   Lobos-Catalan P Jimenez-Castillo M( 2014) Different patterns of biomass allocation of mature and sapling host tree in response to liana competition in the southern temperate rainforest. Austral Ecol  39: 677– 85. Google Scholar CrossRef Search ADS   Luo YJ Wang XK Zhang XQet al.  ( 2012) Root:shoot ratios across china’s forests: forest type and climatic effects. For Ecol Manag  269: 19– 25. Google Scholar CrossRef Search ADS   Makita N Hirano Y Mizoguchi Tet al.  ( 2011) Very fine roots respond to soil depth: biomass allocation, morphology, and physiology in a broad-leaved temperate forest. Ecol Res  26: 95– 104. Google Scholar CrossRef Search ADS   McCarthy MC Enquist BJ( 2007) Consistency between an allometric approach and optimal partitioning theory in global patterns of plant biomass allocation. Funct Ecol  21: 713– 20. Google Scholar CrossRef Search ADS   McConnaughay KDM Coleman JS( 1999) Biomass allocation in plants: ontogeny or optimality? A test along three resource gradients. Ecology  80: 2581– 93. Google Scholar CrossRef Search ADS   Ostle NJ Smith P Fisher Ret al.  ( 2009) Integrating plant-soil interactions into global carbon cycle models. J Ecol  97: 851– 63. Google Scholar CrossRef Search ADS   Pan YD Birdsey RA Fang JYet al.  ( 2011) A large and persistent carbon sink in the world’s forests. Science  333: 988– 93. Google Scholar CrossRef Search ADS PubMed  Physical Geography in China Editorial Board of Chinese Academy of Sciences( 1985) Physical Geography in China: Climatology . Beijing: Science Press, 1– 161. Piao SL Fang JY Zhu Bet al.  ( 2005) Forest biomass carbon stocks in china over the past 2 decades: estimation based on integrated inventory and satellite data. J Geophys Res  110:G01006. Poorter H Nagel O( 2000) The role of biomass allocation in the growth response of plants to different levels of light, CO2, nutrients and water: a quantitative review. Aust J Plant Physiol  27: 595– 607. Google Scholar CrossRef Search ADS   Post WM Zobler L( 2000) Global Soil Types, 0.5-Degree Grid (Modified Zobler) . Data set. Oak Ridge, Tennessee, USA: Oak Ridge National Laboratory Distributed Active Archive Center. http://www.daac.ornl.gov ( 1 October 2015, date last accessed). Purves D Pacala S( 2008) Predictive models of forest dynamics. Science  320: 1452– 3. Google Scholar CrossRef Search ADS PubMed  Qin Y Yi SH Chen JJet al.  ( 2015) Effects of gravel on soil and vegetation properties of alpine grassland on the Qinghai-Tibetan plateau. Ecol Eng  74: 351– 5. Google Scholar CrossRef Search ADS   R Development Core Team( 2016)R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org/ (15 February 2016, date last accessed). Read J Stokes A( 2006) Plant biomechanics in an ecological context. Am J Bot  93: 1546– 65. Google Scholar CrossRef Search ADS PubMed  Reich PB Luo YJ Bradford JBet al.  ( 2014) Temperature drives global patterns in forest biomass distribution in leaves, stems, and roots. Proc Natl Acad Sci USA  111: 13721– 6. Google Scholar CrossRef Search ADS PubMed  Ren W Tian HQ Tao Bet al.  ( 2011) Impacts of tropospheric ozone and climate change on net primary productivity and net carbon exchange of China’s forest ecosystems. Glob Ecol Biogeogr  20: 391– 406. Google Scholar CrossRef Search ADS   Roa-Fuentes LL Campo J Parra-Tabla V( 2012) Plant biomass allocation across a precipitation gradient: an approach to seasonally dry tropical forest at Yucatan, Mexico. Ecosystems  15: 1234– 44. Google Scholar CrossRef Search ADS   Schall P Lodige C Beck Met al.  ( 2012) Biomass allocation to roots and shoots is more sensitive to shade and drought in European beech than in Norway spruce seedlings. For Ecol Manag  266: 246– 53. Google Scholar CrossRef Search ADS   Shinozaki Y Shaw R Lichtin NN( 1964) Reactions of active nitrogen with organic substrates 2. Molecular origins of products of reaction with propene. J Am Chem Soc  86: 341. Google Scholar CrossRef Search ADS   Shipley B Meziane D( 2002) The balanced-growth hypothesis and the allometry of leaf and root biomass allocation. Funct Ecol  16: 326– 31. Google Scholar CrossRef Search ADS   Sun J Cheng GW Fan JH ( 2013a) Soil respiration in response to short-term nitrogen addition in an alpine steppe in northern Tibet. Pol J Ecol  61: 655– 63. Sun J Cheng GW Li WP ( 2013b) Meta-analysis of relationships between environmental factors and aboveground biomass in the alpine grassland on the Tibetan plateau. Biogeosciences  10: 1707– 15. Google Scholar CrossRef Search ADS   Sun J Wang HM( 2016) Soil nitrogen and carbon determine the trade-off of the above- and below-ground biomass across alpine grasslands, Tibetan plateau. Ecol Indic  60: 1070– 6. Google Scholar CrossRef Search ADS   Thornley JH( 1972) Balanced quantitative model for root–shoot ratios in vegetative plants. Ann Bot  36: 431. Google Scholar CrossRef Search ADS   Vandenwollenberg AL( 1977) Redundancy analysis an alternative for canonical correlation analysis. Psychometrika  42: 207– 19. Google Scholar CrossRef Search ADS   Wang LM Li LH Chen Xet al.  ( 2014) Biomass allocation patterns across China’s terrestrial biomes. PLOS ONE  9:e93566. Wang XP Fang JY Zhu B( 2008) Forest biomass and root-shoot allocation in northeast China. For Ecol Manag  255: 4007– 20. Google Scholar CrossRef Search ADS   Warton DI Wright IJ Falster DSet al.  ( 2006) Bivariate line-fitting methods for allometry. Biol Rev  81: 259– 91. Google Scholar CrossRef Search ADS PubMed  Wolf A Field CB Berry JA( 2011) Allometric growth and allocation in forests: a perspective from fluxnet. Ecol Appl  21: 1546– 56. Google Scholar CrossRef Search ADS PubMed  Yang YH Fang JY Ji CJet al.  ( 2009) Above- and belowground biomass allocation in Tibetan grasslands. J Veg Sci  20: 177– 84. Google Scholar CrossRef Search ADS   Yang YH Fang JY Ma WHet al.  ( 2010) Large-scale pattern of biomass partitioning across China’s grasslands. Glob Ecol Biogeogr  19: 268– 77. Google Scholar CrossRef Search ADS   Yang YH Luo YQ( 2011) Isometric biomass partitioning pattern in forest ecosystems: evidence from temporal observations during stand development. J Ecol  99: 431– 7. Google Scholar CrossRef Search ADS   Zhang H Song TQ Wang KLet al.  ( 2016) Influences of stand characteristics and environmental factors on forest biomass and root-shoot allocation in southwest China. Ecol Eng  91: 7– 15. Google Scholar CrossRef Search ADS   Zhao JF Yan XD Guo JPet al.  ( 2012) Evaluating spatial-temporal dynamics of net primary productivity of different forest types in northeastern China based on improved FORCCHN. PLOS ONE  7:e48131. Zhou GS Wang YH Jiang YLet al.  ( 2002) Estimating biomass and net primary production from forest inventory data: a case study of China’s larix forests. For Ecol Manag  169: 149– 57. Google Scholar CrossRef Search ADS   Zhu HZ( 2006) Chinese forest remote sensing classification and carbon density changed pattern based on ecological process parameters. Ph.D. Thesis. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. © The Author(s) 2017. Published by Oxford University Press on behalf of the Institute of Botany, Chinese Academy of Sciences and the Botanical Society of China. All rights reserved. For permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Plant Ecology Oxford University Press

Divergent biomass partitioning to aboveground and belowground across forests in China

Loading next page...
 
/lp/ou_press/divergent-biomass-partitioning-to-aboveground-and-belowground-across-0e0PR9qsdv
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of the Institute of Botany, Chinese Academy of Sciences and the Botanical Society of China. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
1752-9921
eISSN
1752-993X
D.O.I.
10.1093/jpe/rtx021
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Belowground to aboveground biomass (BGB/AGB) ratio is a highly valued parameter of the terrestrial carbon cycle and productivity. However, it remains far from clear whether plant biomass partitioning to aboveground and belowground is isometric (equal partitioning) or allometric (unequal partitioning) at community levels and what factors are necessary in order to regulate the partitioning. This study aimed to comprehensively find out the patterns of biomass partitioning and their regulatory factors across forests in China. Methods The data of AGB and BGB were compiled from 1542 samples for communities across forests in China. Standardized major axis regression was conducted to examine whether AGB and BGB were allocated isometrically or allometrically at a community level. Redundancy analysis was used to analyze the relationships of BGB/AGB ratio with climatic factors and soil properties. Important Findings We found that the slopes of the relationship between logAGB and logBGB were not always comparable to 1.0 (isometric allocation) at community levels, including primary forest, secondary forest, and planted forest. Meanwhile, samples in clay, loam, and sand soil types also presented the same phenomenon. Furthermore, the radically different allocations of AGB and BGB were found in northern and southern China. Environmental factors totally explained 3.86% of the variations in the BGB/AGB ratio at the community level, which include the mean annual precipitation, mean annual temperature, potential water deficit index, soil carbon content, soil nitrogen content, soil clay, soil loam, soil sand, soil pH, and soil bulk density. In addition, the environmental factors also have effects on the BGB/AGB ratio in other categories. The patterns revealed in this study are helpful for better understanding biomass partitioning and spreading the carbon circle models. biomass partitioning, isometric partitioning, BGB/AGB ratio, environmental factors, climatic zone, Chinese forest INTRODUCTION The belowground to aboveground biomass (BGB/AGB) ratio is a variable reflecting a plant’s response and adaption strategies to environmental stress and is also an important parameter for the terrestrial carbon cycle (Jackson et al. 2000; Wolf et al. 2011). Due to limited study of belowground part with great difficulty and high cost, biomass partitioning has not been well constrained in models, which induces large uncertainties in simulating the global carbon cycle and its responses to climate change (Wolf et al. 2011). It remains contentious whether biomass partitioning to belowground and aboveground is isometric or allometric, although it has been extensively examined at community levels in previous studies (Enquist 2002; McCarthy et al. 2007; McConnaughay et al. 1999; Shipley et al. 2002). Even less clear is how much of an impact can environmental factors have. Biomass partitioning is usually quantified as a slope of log-transformed aboveground biomass (logAGB) versus log-transformed belowground biomass (logBGB), which can be statistically identical to either 1 (isometric biomass partitioning) or not (allometric biomass partitioning). Regarding the isometric biomass partitioning, Cheng et al. (2007) reported that the relationships between logAGB and logBGB follow an isometric relationship across forested communities in China. Further, BGB/AGB ratio also did not have any significant linear trend with stand age across various forest ecosystems (Yang et al. 2011). Meanwhile, the isometric biomass partitioning has been verified in other ecosystems, for instance, alpine grasslands (Yang et al. 2009) and all grasslands in China (Yang et al. 2010). Nevertheless, Wang et al. (2008) found that the BGB/AGB ratio had negative relationships with water availability, stand age, height and volume, and hold, which had significant effects on the climate and ontogeny of biomass allocation. In addition, soil nutrients play important roles in biomass allocation; the lower phosphorus availability results in the higher allocation of belowground biomass and production (Castaneda-Moya et al. 2013), and light availability also affects plant biomass allocation (Larcher et al. 2012). These partitioning patterns can be seen as a result of increased allocation by plants of photosynthates to roots in low resource environments. This may also give rise to an increase in water and nutrient absorption for survival. (Lobos-Catalan et al. 2014; Sun et al. 2016). Though there is extensive knowledge on the partitioning patterns of AGB and BGB there is still need for integrated research at community level that investigates the influence of abiotic factors on biomass partitioning. To our knowledge, forests contribute largely to terrestrial ecosystem carbon uptake and thus are crucial components in regulating global and regional carbon cycles (Pan et al. 2011; Piao et al. 2005; Zhao et al. 2012). A wide range of forests can be found in China, from conifer to tropical forests along with other forest types (Fang et al. 2012) with the high mountain forests in the Tibetan Plateau exceptionally diverse when compared to other countries in the world. This diverse and varied range of forests with large spatial variations in biomass, climate and soil texture along with different factors for biomass partitioning offer natural experimental sites for research on structure and function of forest ecosystems. Forests in China extensively represent all forest types in the world from conifer forest to tropical forest as well as other types (Fang et al. 2012), whose diversity in China is even more than that in other countries of the world for high mountain forests in the Tibetan Plateau of China. Thus, China owns natural experimental fields for studying the structure and function of forest ecosystems and the controlling factors for biomass partitioning at the regional scale for the large spatial variations of biomass, climate and soil texture across all types of forests (Luo et al. 2012; Ren et al. 2011; Zhou et al. 2002). In this study, we aimed to examine variations in biomass partitioning across different forest ecosystems in China and their impact factors. The specific objectives were: (i) to investigate whether AGB scales isometrically or allometrically with BGB at a community level; (ii) to compare biomass partitioning patterns among different climatic zones; and (iii) to comprehensively examine the impacts of climate factors and soil properties, on biomass partitioning. MATERIAL AND METHODS Data compilation Biomass data for this study also utilized data that was collected from previous studies (Wang et al. 2014), with the total number of samples accounting to 1542 in the final database from the range of forests in China (see online supplementary Table S1). The following criteria were used to screen and filter the documents: (i) inclusion of both coarse and fine roots in BGB; (ii) all the samples were collected from mature forests; and (iii) the community biomass was calculated using the ‘standard tree’ method (Hui et al. 2014). Our database consisted of geographic location (longitude and latitude) and two target variables (AGB and BGB). Climatic variables [i.e. annual mean temperature, annual mean precipitation, potential water deficit index] and soil properties [i.e. soil nitrogen density (N), pH, soil carbon density (C), soil bulk density (bulk) and soil texture] were collected to analyze the relationships of environmental factors with AGB/BGB and their ratio in each study paper. The soil variables were collected from the data set of Zhu (2006) and Post (2000). Study area The study sites include all the forest types in China, which span a wide range of environmental conditions (Gao et al. 2013). The study area covered from 80.24° to 130.89°E in longitude and ranged from 41.28° to 44.443°N in latitude (Fig. 1). According to the category of physical geography in China (Physical Geography in China Editorial Board of Chinese Academy of Sciences 1985), climatic zones in China include the northern and southern zones. Meanwhile, the B line is 0°C isothermal line in January, and 800 mm precipitation line (Fig. 1). Figure 1: View largeDownload slide spatial distribution of sampling sites. A total of 1543 samples of aboveground biomass (AGB) and belowground biomass (BGB) were compiled from the wide ranging forest types in China. The red solid circles, yellow solid circles and green solid circles represent primary forest, planted forest and secondary forest, respectively. Some plots are not visible as there are instances of plot overlap. Line A is the boundary between the monsoon and non-monsoon and 400 mm of precipitation line. Line B is 0°C isothermal line in January and 800 mm precipitation line. Line C is the precipitation and topography line (Kunlun Mountains, Qilian Mountains, Hengduan Mountains). Figure 1: View largeDownload slide spatial distribution of sampling sites. A total of 1543 samples of aboveground biomass (AGB) and belowground biomass (BGB) were compiled from the wide ranging forest types in China. The red solid circles, yellow solid circles and green solid circles represent primary forest, planted forest and secondary forest, respectively. Some plots are not visible as there are instances of plot overlap. Line A is the boundary between the monsoon and non-monsoon and 400 mm of precipitation line. Line B is 0°C isothermal line in January and 800 mm precipitation line. Line C is the precipitation and topography line (Kunlun Mountains, Qilian Mountains, Hengduan Mountains). Data analysis To examine the allocation pattern (isometric or allometric) of AGB and BGB across different field sites and climatic zones at community levels the standardized major axis (SRMA) regression was conducted (Enquist et al. 2002). The regression relationship of AGB = βBGBα or LogAGB = αLogBGB + Logβ was used to describe the relationship between AGB and BGB, where α is the slope and Logβ is the y-intercept (Li et al. 2006). The Standardized Major Axis Tests & Routines Version 2.0 software package was used to determine the scaling slope and y-intercept of the function (Warton et al. 2006). The relationship between AGB and BGB was considered isometric if a 95% confidence interval (CI) of the scaling slope covered 1.0 and vice versa. Relationship among multiple interacting variables can be effectively evaluated using redundancy analysis (RA; Chen et al. 2016; Vandenwollenberg 1977); thus, the RA was used to analyze the relationships of environmental factors with BGB/AGB, and environmental factors were partitioned into two explanatory variable groups (climatic factors and soil factors). This process was operated calling program ‘vegan’ in the R software package (R Development Core Team, 2016). RESULTS Variations in BGB/AGB ratio At the community level the BGB/AGB ratio showed a mean value of 0.24 and variations ranging from 0.02 to 1.96 across all forest types (Fig. 2). The mean values of BGB/AGB ratio in clay, loam and sand soil types were 0.23, 0.23 and 0.26, respectively, while that in planted, primary and secondary forest BGB were 0.24, 0.25 and 0.27 separately. The maximum (ranging from 0.03 to 1.96) and minimum (ranging from 0.05 to 0.50) of variation were found in sand and clay soil types. Figure 2: View largeDownload slide the frequency distribution of belowground biomass (BGB)/aboveground biomass (AGB) in clay, loam, and sand soil types and in planted forest, primary forest, secondary forest and all samples in China at the community level. Figure 2: View largeDownload slide the frequency distribution of belowground biomass (BGB)/aboveground biomass (AGB) in clay, loam, and sand soil types and in planted forest, primary forest, secondary forest and all samples in China at the community level. Biomass partitioning patterns in different forest types BGB and AGB had a relationship being characterized by the log-transformed linear function of LogAGB = 0.71 + 0.96LogBGB (R2 = 0.84, P < 0.01) in planted forest. The fitting model had a slope of 0.96 at 95% CI between 0.93 and 0.98 (Fig. 3A), which followed allometric allocation, and slope was significantly different from 1. In primary forest, the relationship between BGB and AGB was characterized by the log-transformed linear function of LogAGB = 0.65 + 0.98LogBGB (R2 = 0.83, P > 0.05). The slope of the fitting model was 0.98, with a 95% CI of 0.94–1.02 (Fig. 3B), which followed isometric allocation. BGB showed a linear relationship with AGB in secondary forests after it was log transformed and characterized in function of LogAGB = 0.69 + 0.93LogBGB (R2 = 0.84, P > 0.05). The slope of the fitting model was 0.83, with a 95% CI of 0.84–1.02 (Fig. 3C), which followed isometric allocation. The log-transformed linear function of LogAGB = 0.69 + 0.96LogBGB was used to characterize the relationship between BGB and RGB, across different forest types. The fitting model had a slope of 0.96, with a 95% CI of 0.94–0.98 (Fig. 3D) that indicated a allometric allocation pattern (P < 0.001). Figure 3: View largeDownload slide in the planted forests the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level (A), primary forest (B), secondary forest (C) and all samples in China (D). The standardized major axis regression was used for all panels in the graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 3: View largeDownload slide in the planted forests the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level (A), primary forest (B), secondary forest (C) and all samples in China (D). The standardized major axis regression was used for all panels in the graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Biomass partitioning patterns in different soil types In clay types, the relationship between BGB and AGB was characterized by the log-transformed linear function of LogAGB = 0.63 + 1.03LogBGB (R2 = 0.90, P > 0.05). The slope of the fitting model was 1.03, with a 95% CI of 0.99–1.06 (Fig. 4). In loam soil types, the slope of the relationship between log AGB and log BGB was 1.02 with 95% CIs of 0.99–1.05. Thus, the biomass partitioning patterns in clay and loam soil types followed isometric allocation. Forests found in sandy soils showed opposite results with the relationship between BGB and AGB characterized by the log-transformed linear function of LogAGB = 0.62 + 1.02LogBGB (R2 = 0.79, P < 0.001), whose slope in the fitting model was 1.10, with a 95% CI of 1.04–1.17. Figure 4: View largeDownload slide under different soil types like clay, loan and sand the relationship between aboveground biomass (AGB) and belowground biomass (BGB). The standardized major axis regression was used for the graph. AGB and BGB was log 10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 4: View largeDownload slide under different soil types like clay, loan and sand the relationship between aboveground biomass (AGB) and belowground biomass (BGB). The standardized major axis regression was used for the graph. AGB and BGB was log 10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Biomass partitioning patterns in different climatic zones The relationships between BGB and AGB in different climate zones (Fig. 5). Slopes and the 95% CIs of fitting models were 0.99 (0.96–1.02) in northern China (Fig. 5A) and 1.05 (1.02–1.07) in southern China, respectively. Equations’ slopes had non-significant difference with 1 in northern China (Fig. 5B; P > 0.05), while significant difference the slope from 1 in southern China (P< 0.001), respectively. Figure 5: View largeDownload slide under different climate zones in China the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level. (A and B) Forest in northern China and southern China. Standardized major axis regression is used for all panels in graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Figure 5: View largeDownload slide under different climate zones in China the relationship between aboveground biomass (AGB) and belowground biomass (BGB) at the community level. (A and B) Forest in northern China and southern China. Standardized major axis regression is used for all panels in graph. AGB and BGB were all log10-transformed. P < 0.05 indicates the fitting line is significantly different with 1:1 line. Responses of BGB/AGB ratio to environmental factors The RA was used to further estimate the causal relationship among the interactive variables (Fig. 6). Result showed that all the environmental variables explained 3.86% of the variation in the BGB/AGB ratio of all forests in China (Fig. 6G). The effects of these environmental factors on the BGB/AGB ratio were as follows: climate factors (1.30%); soil factors (2.77%) and coupling effect (−0.21%). Meanwhile, the environmental factors have a small effect on the BGB/AGB ratio in other categories (Fig. 6A–B, D–F, and H–I). Figure 6: View largeDownload slide the contributions of environmental factors to BGB/AGB via redundancy analysis is presented in Venn diagrams. Contributions of climate factors and soil factors to the BGB/AGB in planted forest (A), primary forest (B), secondary forest (C), clay forest (D), loam forest (E), sand forest (F), forest in China (G), forest in northern China (H) and forest in southern China (I). Figure 6: View largeDownload slide the contributions of environmental factors to BGB/AGB via redundancy analysis is presented in Venn diagrams. Contributions of climate factors and soil factors to the BGB/AGB in planted forest (A), primary forest (B), secondary forest (C), clay forest (D), loam forest (E), sand forest (F), forest in China (G), forest in northern China (H) and forest in southern China (I). Mechanisms of observed differences in the root–shoot relationship Based on the above analysis, we developed a framework to explain observed differences of the root–shoot relationship (Fig. 7). Biomass allocation of plants could be influenced by environment stress. The scaling slope should be equal to 1 (isometric allocation) under non-stressful conditions. However, the scaling slope of plant can be flexible, which may vary from the minimum to the maximum and is regulated by environmental factors. The response of biomass allocation sometimes might produce ‘redundancy’ theoretically once it falls into ‘tolerance’ area. Plants were able to adjust the ratio of root–shoot within a certain range to adapt to the variation of environments. Figure 7: View largeDownload slide the mechanisms suggested explaining observed differences in the variation of BGB/AGB. Figure 7: View largeDownload slide the mechanisms suggested explaining observed differences in the variation of BGB/AGB. DISCUSSION Divergent biomass partitioning across forests in China Theoretically, optimal partitioning theory hold a view that plants prefer to partition more biomass to the organ to acquire the most limiting resource (Bloom et al. 1985; Thornley 1972). Contrary to this belief, the allometric biomass partitioning theory predicts that to minimize hydrodynamic resistance and transport duration and to maximize photosynthetic harvesting capacity and resource transport, plant may redistribute biomass (Enquist and Niklas 2002; McCarthy and Enquist 2007). However, in reality, both allometric (Roa-Fuentes et al. 2012) and isometric (Cheng and Niklas 2007; Poorter et al. 2000; Yang and Luo 2011) partitioning were documented in experimental studies at the community level. In this study, there is divergent biomass partitioning to aboveground and belowground across forests in China. At the community level, the co-existing species may have trade-off effects with each other to maximize the resource use efficiency (Kneitel et al. 2004), and the partitioning varies among different climate zones at community levels. We found that the northern climate zone (Fig. 5A) followed isometric pattern (P > 0.05), but the southern climate zone (Fig. 5B) was allometric allocation (slope = 1.05). In addition, biomass allocation in sand soils was statistically allometric but was isometric in clay and loam soils (Fig. 4). In sandy soil, high silt content will decrease macropores and increase bulk density (Archer et al. 2002), with a consequent limitation in root growth (Dostal et al. 2005). Our findings documented that different climate zones and soil types have different biomass partitioning patterns and indicate that climate elements and soil texture have impacts on biomass partitioning in forests. Furthermore, the phenomenon also was proved by RA (Fig. 6); 3.86% of the variation in BGB/AGB ratio was explained by environmental factors; and climate factors, soil factors and coupling effect were 1.30%, 2.77% and −0.21%, respectively. Meanwhile, the environmental factors explaining the variation in BGB/AGB ratio were 4.06%, 8.24%, 37.16%, 9.59%, 5.39%, 11.89%, 8.43% and 6.32% in planted forest, primary forest, secondary forest, clay forest, loam forest, sand forest, forest in northern China and forest in southern China. At large spatial and temporal scales precipitation and temperature act as limiting factors for the growth and distribution of vegetation (Sun et al. 2013b). Our result is consistent with previous studies that found temperature (Reich et al. 2014) and precipitation (Roa-Fuentes et al. 2012) was a driver for forest biomass distribution in aboveground and belowground biomass. Soil characteristics may be important factors influencing BGB/AGB besides climate factors (Becknell et al. 2014). Plants will allocate more biomass into roots for more uptake of soil nutrient (Gleason et al. 2011) or available water (Sun et al. 2013b), some local studies have shown changing allocation of BGB/AGB, regulated by soil characteristics [e.g. soil carbon (Sun and Wang 2016) and nitrogen (Makita et al. 2011; Sun et al. 2013a)]. Meanwhile, soil texture was significantly correlated with vegetation biomass in forest ecosystems, and the same result also was found in alpine grassland (Qin et al. 2015). In our study, our data also indicated BGB/AGB ratio was affected by soil factors (Fig. 6). This suggests that environmental factors have an impact on biomass partitioning in different categories across Chinese forests. Generally, this study more comprehensively revealed forest biomass partitioning in different forest types, soil types, and climate zones. The findings indicated that our results do not agree with the isometric allocation hypothesis regarding the relationship between AGB and BGB across Chinese forests. The BGB/AGB varied with the change of ambient conditions, and the biomass allocation also was governed by self-regulation (Wang et al. 2008; Zhang et al. 2016). Plants may have self-regulation capabilities within a certain range of environmental changes or disturbances in terms of biomass partitioning (Read et al. 2006). Once this self-regulation exceeds certain ‘minimum’ or ‘maximum’ limits, biomass would be over-partitioned to some extent (Kays et al. 1974). Hence, trade-off between biomass partitioning between aboveground and belowground is driven by both external ‘environmental filtering’ and internal adaptation strategies (Fig. 7). Implications for ecosystem modeling Assuming that a fixed fraction of the carbon assimilation is allocated to each organ makes it convenient to model carbon allocation (Ostle et al. 2009). Thus, to estimate future carbon budgets, dynamic global vegetation models have been used with fixed or only water dependent carbon allocation according to pipe theory, which was proposed based on the relationship between leaf water use and stem transport capacity (Ostle et al. 2009; Shinozaki et al. 1964). This leads to the large uncertainty in predicting the future global forest carbon balance (Purves et al. 2008). The simplification of biomass partitioning in models is due to a lack of empirical evidence on biomass partitioning and the difficulty in interpreting the available information under a wide range of conditions (Franklin et al. 2012). Meanwhile, most research about the carbon circle mostly based on model estimates, while the experimental data are penurious at large scale. With variations in climate and soil conditions, the different biomass partitioning patterns with climate zones and environmental factors were found. The results provide both empirical evidence and the environmental drivers for the biomass allocation in forests. The pipe model relationship is mainly governed by the internal relationship between plant organs for water; it is thus less invariant across environment and can’t be used to simulate effects of resource availability and climate change on C allocation in trees (Franklin et al. 2012). The results obtained show the implications on the development of allocation models. First, data to numerically parameterize allocation models are provided by allocation patters in different climate zones. Second, the controlling factors and their relationship with the biomass partitioning provide windows that are to be described mathematically in allocation models. Third, patterns revealed in the study provide a benchmark with which to evaluate and benchmark those quantitative predictions. Limitations of the current study Biomass partitioning is both determined and driven by environmental conditions and ontogeny (Schall et al. 2012, Wang et al. 2008). In the present study, we analyzed the relationships of environmental factors with BGB/AGB; as a matter of fact, the plants were able to self-regulate the ratio of BGB/AGB (roots, stem, and leaves) to adapt to an environment. In addition, a conceptual model (Fig. 7) may be a useful pathway to illustrate biomass allocation of plants to aboveground and belowground, and the allocated process is achieved through signal transduction to morphological or physiological response. The theory proposes that the sum of all modular responses of a plant to its local conditions along with the interaction effects due to integration gives the response of a plant to its environment. According to de Kroon et al. 2005 there may be rules that are seen as evolving traits targeted by natural selection—local response rules to environmental variation and the modular interaction rules. Consequently, the mechanism of self-regulation needs to be further explored in the future. CONCLUSIONS Synthesizing biomass partitioning patterns across forests in China, this study demonstrated that there were no general rules on biomass partitioning in the different soil types, forest types, and climate zones. Environmental factors (climate factors and soil factors) have effects on biomass partitioning. The results suggest that biomass partitioning is governed by external environmental factors and internal plant self-regulation. Our work provides useful constraints on biomass partitioning parameters, which may reduce the uncertainties in predicting the global carbon cycle. SUPPLEMENTARY MATERIAL Supplementary data are available at Journal of Plant Ecology online. FUNDING This study was jointly supported by CAS Strategic Priority Research Program (grant no. XDA05050702), the National Natural Science Foundation of China (31420103917, 31290221, 41661144045), the Thousand Youth Talents Plan Project, and the Open Fund of Key Laboratory of Ecosystem Network Observation and Modelling. ACKNOWLEDGEMENTS We thank Dr Dafeng Hui for his helpful comments that had improved this manuscript greatly. Meanwhile, we appreciate two anonymous reviewers for their professional and useful suggestions regarding our previous manuscript. Conflict of interest statement. None declared. REFERENCES Archer NAL Quinton JN Hess TM( 2002) Below-ground relationships of soil texture, roots and hydraulic conductivity in two-phase mosaic vegetation in south-east Spain. J Arid Environ  52: 535– 53. Google Scholar CrossRef Search ADS   Becknell JM Powers JS( 2014) Stand age and soils as drivers of plant functional traits and aboveground biomass in secondary tropical dry forest. Can J For Res  44: 604– 13. Google Scholar CrossRef Search ADS   Bloom AJ Chapin FS Mooney HA( 1985) Resource limitation in plants—an economic analogy. Annu Rev Ecol Syst  16: 363– 92. Google Scholar CrossRef Search ADS   Castaneda-Moya E Twilley RR Rivera-Monroy VH( 2013) Allocation of biomass and net primary productivity of mangrove forests along environmental gradients in the Florida coastal everglades, USA. For Ecol Manag  307: 226– 41. Google Scholar CrossRef Search ADS   Chen YL Ding JZ Peng YFet al.  ( 2016) Patterns and drivers of soil microbial communities in tibetan alpine and global terrestrial ecosystems. J Biogeogr  43: 2027– 39. Google Scholar CrossRef Search ADS   Cheng DL Niklas KJ( 2007) Above- and below-ground biomass relationships across 1534 forested communities. Ann Bot  99: 95– 102. Google Scholar CrossRef Search ADS PubMed  de Kroon H Huber H Stuefer JFet al.  ( 2005) A modular concept of phenotypic plasticity in plants. New Phytol  166: 73– 82. Google Scholar CrossRef Search ADS PubMed  Dostal P Breznova M Kozlickova Vet al.  ( 2005) Ant-induced soil modification and its effect on plant below-ground biomass. Pedobiologia  49: 127– 37. Google Scholar CrossRef Search ADS   Enquist BJ( 2002) Universal scaling in tree and vascular plant allometry: toward a general quantitative theory linking plant form and function from cells to ecosystems. Tree Physiol  22: 1045– 64. Google Scholar CrossRef Search ADS PubMed  Enquist BJ Niklas KJ( 2002) Global allocation rules for patterns of biomass partitioning in seed plants. Science  295: 1517– 20. Google Scholar CrossRef Search ADS PubMed  Fang JY Wang XP Liu YNet al.  ( 2012) Multi-scale patterns of forest structure and species composition in relation to climate in northeast china. Ecography  35: 1072– 82. Google Scholar CrossRef Search ADS   Franklin PA Bartels B( 2012) Restoring connectivity for migratory native fish in a New Zealand stream: effectiveness of retrofitting a pipe culvert. Aquat Conserv  22: 489– 97. Google Scholar CrossRef Search ADS   Gao YH Liu XX Min CCet al.  ( 2013) Estimation of the northsouth transect of eastern China forest biomass using remote sensing and forest inventory data. Int J Remote Sens  34: 5598– 610. Google Scholar CrossRef Search ADS   Gleason SM Read J Ares A( 2011) Biomass allocation and phosphorus economics of rain-forest seedlings: effects of fertilization and radiation on soil specialists and soil generalists. J Trop Ecol  27: 147– 61. Google Scholar CrossRef Search ADS   Hui DF Wang J Shen WJet al.  ( 2014) Near isometric biomass partitioning in forest ecosystems of china. PLOS ONE  9:e86550. Jackson RB Schenk HJ Jobbagy EGet al.  ( 2000) Belowground consequences of vegetation change and their treatment in models. Ecol Appl  10: 470– 83. Google Scholar CrossRef Search ADS   Kays S Harper JL( 1974) Regulation of plant and tiller density in a grass sward. J Ecol  62: 97– 105. Google Scholar CrossRef Search ADS   Kneitel JM Chase JM( 2004) Trade-offs in community ecology: linking spatial scales and species coexistence. Ecol Lett  7: 69– 80. Google Scholar CrossRef Search ADS   Larcher L Boeger MRT Marques MCM( 2012) Biomass allocation and shade tolerance in tree species of the Atlantic forest. Botany  90: 830– 8. Google Scholar CrossRef Search ADS   Li HT Han XG Wu JG( 2006) Variant scaling relationship for mass-density across tree-dominated communities. J Integr Plant Biol  48: 268– 77. Google Scholar CrossRef Search ADS   Lobos-Catalan P Jimenez-Castillo M( 2014) Different patterns of biomass allocation of mature and sapling host tree in response to liana competition in the southern temperate rainforest. Austral Ecol  39: 677– 85. Google Scholar CrossRef Search ADS   Luo YJ Wang XK Zhang XQet al.  ( 2012) Root:shoot ratios across china’s forests: forest type and climatic effects. For Ecol Manag  269: 19– 25. Google Scholar CrossRef Search ADS   Makita N Hirano Y Mizoguchi Tet al.  ( 2011) Very fine roots respond to soil depth: biomass allocation, morphology, and physiology in a broad-leaved temperate forest. Ecol Res  26: 95– 104. Google Scholar CrossRef Search ADS   McCarthy MC Enquist BJ( 2007) Consistency between an allometric approach and optimal partitioning theory in global patterns of plant biomass allocation. Funct Ecol  21: 713– 20. Google Scholar CrossRef Search ADS   McConnaughay KDM Coleman JS( 1999) Biomass allocation in plants: ontogeny or optimality? A test along three resource gradients. Ecology  80: 2581– 93. Google Scholar CrossRef Search ADS   Ostle NJ Smith P Fisher Ret al.  ( 2009) Integrating plant-soil interactions into global carbon cycle models. J Ecol  97: 851– 63. Google Scholar CrossRef Search ADS   Pan YD Birdsey RA Fang JYet al.  ( 2011) A large and persistent carbon sink in the world’s forests. Science  333: 988– 93. Google Scholar CrossRef Search ADS PubMed  Physical Geography in China Editorial Board of Chinese Academy of Sciences( 1985) Physical Geography in China: Climatology . Beijing: Science Press, 1– 161. Piao SL Fang JY Zhu Bet al.  ( 2005) Forest biomass carbon stocks in china over the past 2 decades: estimation based on integrated inventory and satellite data. J Geophys Res  110:G01006. Poorter H Nagel O( 2000) The role of biomass allocation in the growth response of plants to different levels of light, CO2, nutrients and water: a quantitative review. Aust J Plant Physiol  27: 595– 607. Google Scholar CrossRef Search ADS   Post WM Zobler L( 2000) Global Soil Types, 0.5-Degree Grid (Modified Zobler) . Data set. Oak Ridge, Tennessee, USA: Oak Ridge National Laboratory Distributed Active Archive Center. http://www.daac.ornl.gov ( 1 October 2015, date last accessed). Purves D Pacala S( 2008) Predictive models of forest dynamics. Science  320: 1452– 3. Google Scholar CrossRef Search ADS PubMed  Qin Y Yi SH Chen JJet al.  ( 2015) Effects of gravel on soil and vegetation properties of alpine grassland on the Qinghai-Tibetan plateau. Ecol Eng  74: 351– 5. Google Scholar CrossRef Search ADS   R Development Core Team( 2016)R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org/ (15 February 2016, date last accessed). Read J Stokes A( 2006) Plant biomechanics in an ecological context. Am J Bot  93: 1546– 65. Google Scholar CrossRef Search ADS PubMed  Reich PB Luo YJ Bradford JBet al.  ( 2014) Temperature drives global patterns in forest biomass distribution in leaves, stems, and roots. Proc Natl Acad Sci USA  111: 13721– 6. Google Scholar CrossRef Search ADS PubMed  Ren W Tian HQ Tao Bet al.  ( 2011) Impacts of tropospheric ozone and climate change on net primary productivity and net carbon exchange of China’s forest ecosystems. Glob Ecol Biogeogr  20: 391– 406. Google Scholar CrossRef Search ADS   Roa-Fuentes LL Campo J Parra-Tabla V( 2012) Plant biomass allocation across a precipitation gradient: an approach to seasonally dry tropical forest at Yucatan, Mexico. Ecosystems  15: 1234– 44. Google Scholar CrossRef Search ADS   Schall P Lodige C Beck Met al.  ( 2012) Biomass allocation to roots and shoots is more sensitive to shade and drought in European beech than in Norway spruce seedlings. For Ecol Manag  266: 246– 53. Google Scholar CrossRef Search ADS   Shinozaki Y Shaw R Lichtin NN( 1964) Reactions of active nitrogen with organic substrates 2. Molecular origins of products of reaction with propene. J Am Chem Soc  86: 341. Google Scholar CrossRef Search ADS   Shipley B Meziane D( 2002) The balanced-growth hypothesis and the allometry of leaf and root biomass allocation. Funct Ecol  16: 326– 31. Google Scholar CrossRef Search ADS   Sun J Cheng GW Fan JH ( 2013a) Soil respiration in response to short-term nitrogen addition in an alpine steppe in northern Tibet. Pol J Ecol  61: 655– 63. Sun J Cheng GW Li WP ( 2013b) Meta-analysis of relationships between environmental factors and aboveground biomass in the alpine grassland on the Tibetan plateau. Biogeosciences  10: 1707– 15. Google Scholar CrossRef Search ADS   Sun J Wang HM( 2016) Soil nitrogen and carbon determine the trade-off of the above- and below-ground biomass across alpine grasslands, Tibetan plateau. Ecol Indic  60: 1070– 6. Google Scholar CrossRef Search ADS   Thornley JH( 1972) Balanced quantitative model for root–shoot ratios in vegetative plants. Ann Bot  36: 431. Google Scholar CrossRef Search ADS   Vandenwollenberg AL( 1977) Redundancy analysis an alternative for canonical correlation analysis. Psychometrika  42: 207– 19. Google Scholar CrossRef Search ADS   Wang LM Li LH Chen Xet al.  ( 2014) Biomass allocation patterns across China’s terrestrial biomes. PLOS ONE  9:e93566. Wang XP Fang JY Zhu B( 2008) Forest biomass and root-shoot allocation in northeast China. For Ecol Manag  255: 4007– 20. Google Scholar CrossRef Search ADS   Warton DI Wright IJ Falster DSet al.  ( 2006) Bivariate line-fitting methods for allometry. Biol Rev  81: 259– 91. Google Scholar CrossRef Search ADS PubMed  Wolf A Field CB Berry JA( 2011) Allometric growth and allocation in forests: a perspective from fluxnet. Ecol Appl  21: 1546– 56. Google Scholar CrossRef Search ADS PubMed  Yang YH Fang JY Ji CJet al.  ( 2009) Above- and belowground biomass allocation in Tibetan grasslands. J Veg Sci  20: 177– 84. Google Scholar CrossRef Search ADS   Yang YH Fang JY Ma WHet al.  ( 2010) Large-scale pattern of biomass partitioning across China’s grasslands. Glob Ecol Biogeogr  19: 268– 77. Google Scholar CrossRef Search ADS   Yang YH Luo YQ( 2011) Isometric biomass partitioning pattern in forest ecosystems: evidence from temporal observations during stand development. J Ecol  99: 431– 7. Google Scholar CrossRef Search ADS   Zhang H Song TQ Wang KLet al.  ( 2016) Influences of stand characteristics and environmental factors on forest biomass and root-shoot allocation in southwest China. Ecol Eng  91: 7– 15. Google Scholar CrossRef Search ADS   Zhao JF Yan XD Guo JPet al.  ( 2012) Evaluating spatial-temporal dynamics of net primary productivity of different forest types in northeastern China based on improved FORCCHN. PLOS ONE  7:e48131. Zhou GS Wang YH Jiang YLet al.  ( 2002) Estimating biomass and net primary production from forest inventory data: a case study of China’s larix forests. For Ecol Manag  169: 149– 57. Google Scholar CrossRef Search ADS   Zhu HZ( 2006) Chinese forest remote sensing classification and carbon density changed pattern based on ecological process parameters. Ph.D. Thesis. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. © The Author(s) 2017. Published by Oxford University Press on behalf of the Institute of Botany, Chinese Academy of Sciences and the Botanical Society of China. All rights reserved. For permissions, please email: journals.permissions@oup.com

Journal

Journal of Plant EcologyOxford University Press

Published: Jun 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off