- Email: [email protected]
Analyzing the impact of climate and management factors on the productivity and soil carbon sequestration of poplar plantations
Environmental Research 144 (2016) 88–95
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
Analyzing the impact of climate and management factors on the productivity and soil carbon sequestration of poplar plantations Dan Wang a,n, Jiazhi Fan a, Panpan Jing a, Yong Cheng a, Honghua Ruan b a International Center for Ecology, Meteorology and Environment, School of Applied Meteorology, Nanjing University of Information Science and Technology, Nanjing 210044, PR China b Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 9 May 2015 Received in revised form 9 October 2015 Accepted 15 October 2015 Available online 31 October 2015
It is crucial to investigate how climate and management factors impact poplar plantation production and soil carbon sequestration interactively. We extracted above-ground net primary production (ANPP), climate and management factors from peer-reviewed journal articles and analyzed impact of management factor and climate on the mean annual increment (MAI) of poplar ANPP statistically. Previously validated mechanistic model (ED) is used to perform case simulations for managed poplar plantations under different harvesting rotations. The meta-analysis indicate that the dry matter MAI was 6.3 Mg ha 1 yr 1 (n ¼ 641, sd ¼4.9) globally, and 5.1 (n ¼292, sd¼ 4.0), 8.1 (n ¼ 224, sd¼ 4.7) and 4.4 Mg ha 1 yr 1 (n ¼125, sd ¼3.2) in Europe, the US and China, respectively. Poplar MAI showed a significant response to GDD, precipitation and planting density and formed a quadratic relationship with stand age. The low annual production for poplar globally was probably caused by suboptimal water availability, rotation length and planting density. SEM attributes the variance of poplar growth rate more to climate than to management effects. Case simulations indicated that longer rotation cycle significantly increased soil carbon storage. Findings of this work suggests that management factor of rotation cycle alone could have dramatic impact on the above ground growth, as well as on the soil carbon sequestration of poplar plantations and will be helpful to quantify the long-term carbon sequestration through short rotation plantation. The findings of this study are useful in guiding further research, policy and management decisions towards sustainable poplar plantations. & 2015 Elsevier Inc. All rights reserved.
Keywords: Meta-analysis Short rotation forestry Structural equation model Ecosystem Demography model Stand age Planting density
1. Introduction Hybrid poplars are among the most widely cultivated hardwood species for pulp and timber production due to their fast growth rate, high light-use efficiency and photosynthetic capacity, strong tolerance to biophysical stress, ease of vegetative propagation and adaptation to a wide variety of soils (Weih, 2004). There is recognition that fast-growing, high-yield plantations will be important in meeting an increasing global demand for wood and biofuel products to substitute fossil fuels and carbon intensive materials such as steel (Canadian Council of Forest Ministers, 2001; Somerville et al., 2010). Understanding how poplar tree Abbreviations: ANPP, above-ground net primary production; ED, Ecosystem Demography model; GDD, growing degree days; MAI, mean annual increment of ANPP; Pa, annual precipitation; Ps, growing season precipitation; SEM, structural equation model; SOC, soil organic carbon; SRF (C), short rotation forestry (coppice); Tave, annual average temperature; Tmax, annual maximum temperature; Tmin, annual minimum temperature n Corresponding author. E-mail address: [email protected] (D. Wang). http://dx.doi.org/10.1016/j.envres.2015.10.016 0013-9351/& 2015 Elsevier Inc. All rights reserved.
grows has become critically important in recent years as the climate warms and estimates of biomass of forest products (including bioenergy products) and stored forest carbon (50 per cent of forest biomass) are needed. Poplar growth rate vary considerably and are affected by the influence of genetics, climate, and management factors on survival, competition, and vigor of the stand (Mead, 2005). The specific response of poplar growth to climate varies across Populus species (Pan et al., 1997), tree size (Mérian and Lebourgeois, 2011), age (Copenheaver et al., 2011), stand structures (Linares et al., 2010; D'Amato et al., 2013), edaphic or productivity gradients (Orwig and Abrams, 1997; Leonelli et al., 2008), and genetic variability across populations (McLane et al., 2011). Although climate influences tree growth, other management practice such as nitrogen fertilization, harvesting practice and planting density can interact with climate to further affect tree growth (Davis et al., 2012). Past studies linking tree growth with climate have often failed to consider the interacting effects of stand characters and management practices. This likely over-simplifies climate-growth relationships and the potential effect of climate and management practice on both tree- and stand-level productivity. Separating the
D. Wang et al. / Environmental Research 144 (2016) 88–95
impact of climate and management factors on tree growth is useful to guide plantation management in adapting to future climate change conditions. The cultivation of fast-growing woody plants within short rotation forestry (SRF) can provide a potentially important source of wood production, environmental protection and economy income. Woody species managed by short rotation forestry (SRF) have multiple advantages over annual crops which include enhanced biological diversity, greater carbon sequestration, and reduced inputs of labor, pesticides and fertilizers (Hill et al., 2006; Baum et al., 2009; Don et al., 2011). Rotations shorter than 3 years, however, could lead to reduced yields after several rotations due to physiological problems including stump aging and depletion of carbohydrate reserves (Auclair and Bouvarel, 1992) and causes the depletion of nitrogen due to greater consumption of nitrogen than the supply. The maximum biomass productivity is expected with harvest cycles of 3–11 years (Sartori and Lal, 2006). As poplar increased in age, there was a major reduction in nutrient concentrations in the leaves, indicating that nitrogen limitation might become more evident when poplar stand gets older (Wang et al., 2013). Although aboveground woody biomass is the economically important component of SRF ecosystems, enhanced carbon sequestration in roots and soil has a large impact on CO2 mitigation and the ecological benefits of SRF related to conservation, water and soil protection, recreation or climate-change mitigation and adaptation are likely to acquire economic value in the future (Calfapietra et al., 2010). SRF can rapidly accumulate C in stable components such as stems, branches and coarse roots, while at the same time cycling C and nutrients to the soil through more labile litter pools consisting of leaves, twigs and fine roots (Sartori and Lal, 2006; Meiresonne et al., 2006). Observed patterns of soil organic carbon (SOC) dynamics under SRF include short-term losses (Hansen, 1993), long term gains (Hansen, 1993; Makeschin, 1994) and no changes (Ulzen-Appiah et al., 2000). Therefore, determining the optimal rotation cycle will not only be beneficial to poplar yields but also the carbon sequestration level. While the yield potential of poplar has been tested by many on-farm studies and has been modeled and reviewed extensively, to date there have been no quantitative reviews on how climate and management practice affect poplar productivity interactively. And information is scarcer on how rotation cycles affect carbon sequestration in the soil (Anderson-Teixeira et al., 2009). Structural equation modeling is a scientific methodology that aspires to make a strong and explicit connection between empirical data and theoretical ideas (Bollen, 1989; Kline, 2005). It can partition causal influences among multiple variables, allowing the separation effects of different predictors (Grace, 2006). In the present study, we will analyze the effect of climate and management factors on the mean annual increment of poplar ANPP (MAI) intensively and provide guidelines for plantation management to optimize poplar yields and soil carbon sequestration in SRF based on a new, updated quantitative literature review. A structural equation model will be applied to separate the partial effect of climate and management factors on poplar ANPP. We will further run a well-calibrated model on a case site to test the effect of two different rotation cycles on carbon sequestration in poplar SRF. Sustainable forest management practices could maximize carbon sequestration and maintain stand quality and productivity; therefore, the primary goal of this study was to examine the effect of climate and management practice on the growth of poplar across broad geographical ranges. Such quantitative information out of meta-analysis and mechanical models will be used to (1) understand the productivity variability of poplar plantations under different circumstances; (2) identify the key factors impacting poplar MAI and soil carbon sequestration and separating different impacts of climate and management factors on poplar
89
productivity; (3) evaluating the effect of different harvesting cycles on SRF carbon sequestration.
2. Methods The peer-reviewed journal articles used to construct the database for this meta-analysis were obtained by searching the Science Citation Index (SCI) of the Institute of Scientific Information and Chinese Journal Full-text Database (CJFD). The inclusion of journal articles in Chinese ensured the completeness of the growth information for poplar plantation, as poplar plantation area in China is not negligible but journal articles published in Chinese often being excluded in meta-analysis due to languages barriers. The list of articles obtained was subsequently cross-checked with references cited in a large number of review articles and books to ensure the inclusion of all articles containing data relevant for this meta-analysis. Any article published before the end of 2014 that includes the following information was included: (1) biomass of poplar plantation; (2) site location; (3) management information on planting density, stand age, nitrogen fertilization level. Trials with pathogen or disease attacks were excluded from the analysis. In total, 29 peer-reviewed articles from China, 28 from other countries were included in this meta-analysis (Appendix A and B). Meteorological data was obtained from published articles or nearby meteorological stations if available or from LOCCLIM (v. 1.0 FAO, Rome, Italy) for a given site when climate data were not available (Wang et al., 2010). Growing degree days (GDD) were calculated in the growing season with the base temperature of 10 °C. The growing season is defined from last frost in spring to the first frost season in autumn or the date of harvest. We assumed studies conducted at different sites, yields from different treatments (e.g. fertilizer treatments), and different growing seasons were independent. In our analysis, studies in which plants were grown under environmental stresses (e.g., drought, low nutrients, light deficiency and etc.) were excluded. The database of poplar plantation contains climate information including annual average (Tave), maximum (Tmax) and minimum temperature (Tmin), GDD, growing season precipitation (Ps), annual precipitation (Pa), site information (site location, longitude, latitude, altitude, soil type), management routine (planting date, rotation scheme, nitrogen level, planting density, stand age, herbicide), biomass and soil C. For studies with more detailed growth information, carbon storage in leaves, branches, trunk, roots and litter are also included in the database. Graphical data were extracted from the articles using digitizing software (GET DATA GRAPH DIGITIZER v. 2.22). Data were firstly sorted and tested for normality. Squared-root transformed MAI were then analyzed using mixed model analysis of variance. The random effects in the mixed model framework were used to account for site-to-site variability that were not accounted for by the fixed-effect covariates but which potentially caused treatments within sites to not be independent (e.g. soil types, soil micro fauna and etc.). Similarly, species random effects account for the differences that could not be accounted for as fixed effects due to limited and unbalanced replication data (Wang et al., 2010). Stepwise regression was conducted to screen a best model to describe MAI (only two way interactions were considered in the analysis). A mixed model taking individual studies and species as random effects was conducted to test the fixed effects of annual average (Tave), maximum (Tmax) and minimum temperature (Tmin), GDD, growing season precipitation (Ps), annual precipitation (Pa), stand age, planting density, N and their interactions on MAI. Prior to evaluating multivariate models, bivariate relations between all variables in the model were assessed. We examined scatterplots for the presence of outliers, evidence of skewness or kurtosis, and non-linear relationships up to second-order
90
D. Wang et al. / Environmental Research 144 (2016) 88–95
polynomials. The relationship between poplar MAI and SOC and individual climate and management factors was evaluated separately by linear or second-order polynomial regression analysis. Tests of hypotheses were considered significant at P r0.05 level. As a second multivariate approach to examine the relationship of poplar productivity with climate and management factors proportionally, structural equation modeling (SEM) was conducted to separate the effect of climate and management. Maximum likelihood solution procedures and chi-square goodness of fit measures were employed to evaluate model adequacy. Finally, we used a well calibrated and validated model, Ecosystem Demography model (ED2), to simulate the productivity and soil carbon sequestration of short-rotation cropping systems of hybrid poplar (Moorcroft et al., 2001; Wang et al., 2012). Established models of leaf level physiology (Farquhar et al., 1980; Leuning, 1995), allocation (Saldarriaga et al., 1988), biogeochemistry (Parton et al., 1993), land surface biophysics (Walko et al., 2000), and hydrology (Sellers, 1992) are used to simulate plant growth in ED2. Model simulations for poplar SRF were run for 68 years based on a 5-yr or 10-yr rotation cycles at a trial site at Inner Mongolia, China (Hu et al., 2008), with the same parameter sets as used in Wang et al. (2012). ED2 simulates biogeochemical as well as growing processes, which enables ED2 not only predict poplar growth but also the carbon cycling in the soil. The simulation assumed no manipulations of water, N, or P. Meteorological drivers were obtained from cru-NCEP reanalysis data and were sequentially cycled over years from 1981 to 2008 (Viovy and Ciais, 2009). Soil depth and textural class for each site were obtained from the published literature. Forest harvesting in ED2 is treated as a removal of the above ground stem biomass, with roots and leaves entering the litter pool.
3. Results The mean dry matter MAI of poplar in China is 4.4 Mg ha 1 yr 1 (n ¼125, sd ¼3.2) and significantly lower than that in European (5.1 Mg ha 1 yr 1, n ¼ 292, sd ¼4.0) and American countries (8.1 Mg ha 1 yr 1, n ¼224, sd¼ 4.7), and even the average globally (6.3 Mg ha 1 yr 1, n ¼641, sd ¼4.9) (Table 1). For MAI, the variance components of the random effect of individual studies and species were significantly different from 0, suggesting that individual studies and species did differ in their MAI (Table 2). The stepwise regression analysis indicated the set of five factors including Ps, GDD, stand age, planting density, N and some of their interactions significantly affected MAI and explained MAI the best (Table 1). The bivariate models indicate that GDD, Ps and Pa, and planting density had a linear relationship with MAI, and Tmin, Tave, Tmax and stand age formed a quadratic relationship with MAI, while N had no relationship with MAI (Fig 1). In order to separate the effect of climate and management factors on MAI, we conducted a SEM on the observed data. The final SEM fits the data well (X2 ¼12.3, d.f. ¼3, P¼ 0.16). The model indicates that climate and management composite factors significantly influenced MAI. All explanatory variables were significantly correlated with each other, indicating that they were not completely independent. The climate composite factors had a positive correlation and management composite factors a negative correlation with MAI, with path coefficient of 0.63 and 0.32, respectively (Fig. 2). Soil carbon sequestration formed a quadratic relationship with stand age, reaching the lowest value at about 14-yr (Fig. 3A). Root carbon storage, however, reached the highest value at 17-yr (Fig. 3B). We simulated annual wood production and net C sequestration of poplar plantations in a case study located at Inner Mongolia,
Table 1 The mean MAI of poplar plantations in China, Europe, the United State. Regions
MAI
n
sd
Global Europe North America China
6.3 5.1 8.1 4.4
641 292 224 125
4.9 4.0 4.7 3.2
Table 2 Fixed effects from mixed model of stand age, growing degree days (GDD), seasonal precipitation (Ps), planting density, N and their interactions on the mean maximum annual increment of aboveground production, taking individual studies and species as random effects. Effects
F-value
Pr4F
Stand age Planting density N GDD Ps Stand age Planting density Stand age N Stand age Ps Planting density GDD GDD N GDD Ps Planting density N
23.36 5.14 6.96 45.97 41.37 35.74 2.17 0.15 42.85 2.99 8.54 8.59
o 0.001 o 0.05 o 0.001 o 0.001 o 0.001 o 0.001
o 0.001 o 0.001 o 0.001
For all the effects, the dfN (numerator degrees of freedom)¼ 1 and the dfD (denominator degrees of freedom)¼ 643 For the random effect of individual studies: Z¼32.4, P o 0.05. For the random effect of species: Z¼44.6, Po 0.05.
that were completely harvested at either 10- or 5-yr intervals using the ED model (Hu et al., 2008; Wang et al., 2012). The MAI in the first harvesting cycle was comparable to the observed (data not shown). If grown in 10-yr cycles, the model simulations of net C storage (including aboveground biomass and soil carbon accumulation) were negative for the first 22 years and recovery occurred after about 3 harvesting cycles (Fig. 4). With a 5-yr harvesting cycle, the net C storage continued to decline over the 68-yr period (Fig. 4). In the longer harvesting cycle, the poplar plantation was a net sink of C after about 30 years, but in the shorter harvesting cycle, the poplar plantation was a net source of C to the atmosphere for the entire period of time tested.
4. Discussion The global average mean MAI of poplar is lower than the reported harvestable yields of poplar in temperate regions of Europe and North America between 10 and 15 Mg ha 1 yr 1 (Kauter et al., 2003; Van de Walle. et al., 2007; Aylott et al., 2008). This is especially true for the MAI in China, where it is significantly lower than that in Europe, USA and even the global average. Poplar plantation in China has reached more than 700 million hectares and ranked the largest in the world, exceeding the total area of poplar plantation in other countries. Therefore, future research understanding why the poplar MAI in China is not comparable to the global mean is urgently needed. The growth rate of poplar varies among different species/clones and sites and depends on the climate and management factors and their interactions (Table 2 and Fig. 1). The set of five factors including Pr, GDD, stand age, planting density, N and the interactions between stand age and planting density, between GDD and planting density, between GDD and Pr, and between planting density and N significantly affects MAI and explained MAI the best. Temperature factors, including Tmin, Tave and Tmax, all have a
D. Wang et al. / Environmental Research 144 (2016) 88–95
91
Fig. 1. Response of poplar MAI to annual average (Tave), maximum (Tmax) and minimum (Tmin) temperature, GDD, growing season precipitation (Ps), annual precipitation (Pa), stand age, planting density and N. Solid lines indicate least-squares linear regression. Shaded area indicates confidence interval of 0.95. Equations and regression coefficients (R) were shown for significant relations.
significant quadratic relationship with poplar MAI, suggesting that there is an optimal temperature range for poplar growth. GDD and precipitation both affect poplar productivity significantly and have a positive relationship with poplar MAI. Seasonal and annual precipitation has a similar impact on poplar MAI, which is consistent with many previous studies that approved the dependence of poplar growth on water availability (Davis et al., 2012; Wang et al., 2013). Among the studied soil properties, soil texture and moisture condition had the strongest and most reliable influence on mean growth characteristics in 5-year-old hybrid aspen plantations (Tullus et al., 2007). The same tendency has been observed also in North America for poplars in general (Stanturf et al., 2001). GDD can affect poplar MAI through affecting the phenology of flowering time and nutrient translocation (Arora and Boer, 2005),
which in turn will affect the annual aboveground productivity. Fu et al. (2012) found that higher GDD will advance bud burst in spring, put off leaf fall in autumn and extend the growing season. The fact that species/clone choice and climatic factors affects poplar ANPP significantly indicates that managers should select species that tolerate a variety of climates, such as those that can grow over a range of latitudes and altitudes (Kolström et al., 2011; Bellassen and Luyssaert, 2014; Fares et al., 2015). Stand age that maximizes the MAI can be used to inform the time period of rotation cycles. SRF includes single stem production followed either by replanting or by coppicing (Weih, 2004). The studies included in this analysis contain both replanting and coppicing trials; therefore results would vary and in general coppicing will enable a shorter rotation cycle and produce higher
92
D. Wang et al. / Environmental Research 144 (2016) 88–95
Fig. 2. Final structural equation model results (X2 ¼12.3, d.f. ¼ 3, P ¼0.16) with standardized path coefficients. The theoretical construct is represented by latent variables (circle) and all measured variables (rectangles) are represented as effect indicators. The proportion of variance explained (R2) is given beside the response variables in the model. Error terms represent unexplained variance of dependent variables. All pathways were significant (Po 0.05). The goodness of fit for the models was determined using a χ2-test.
yields. Nevertheless, we still found that stand age has a quadratic relationship with MAI, indicating at a stand age of about 17-yr, the MAI reached its maximum (Fig. 1G). The MAI-achieving year calculated here should be integrated with the knowledge on morphological and physiological factors influencing the regeneration and sustainability of the plantation, as well as the economic costs of harvesting and post-harvest processes to evaluate the relative benefits of different rotation cycles (Wang et al., 2013). Furthermore, the current MAI of polar might reach its maximum earlier than the mean MAI. For example, the maximum annual increment of ANPP was reached at the age of 13 years and the MAI at the age of 24 years (Liesebach et al., 1999). The stand age effect on MAI depends on the planting density, and with higher-density stands, MAI reaches maximum in an earlier stage. Planting density had a negative linear relationship with MAI,
mainly caused by the low MAI achieved at very high planting densities, which is consistent with previous studies (DeBell and Harrington, 1997) (Fig. 1H). Ceulemans and Deraedt (1999) stated that for poplar SRF, optimal densities range from 2500 to 10,000 trees ha 1. Even though the canopy and leaf properties of poplar SRF are very dynamic and tend to maximize light interception by the canopy, the fast growth and rapid changes in canopy depth and leaf area during the growing season require an optimal planting density with a dynamic and optimal distribution of foliage area, leaf mass and specific leaf area of the poplar canopy. Thinning can encourage new growth and increase structural diversity, which will increase canopy complexity and promote more efficient use of light and nitrogen (Fares et al., 2015). Many studies have illustrated the beneficial effect of nitrogen treatments on poplar growth and productivity. But due to large differences in fertilization response and maximum growth observed among various hybrid crosses (Heilman and Xie, 1993; Ceulemans and Deraedt, 1999), we could not find a positive effect of N on MAI in this analysis. And contrary to our expectation that N demand might increase as stands age due to N removal from biomass harvesting and leaching out of root zone, the response of poplar MAI to N x stand age treatment is not significant (Table 1). Makeschin and Rehfuess (1994) found no effect in biomass production in fertilized and unfertilized trials. There is a significant N x planting density effect on MAI, suggesting that nitrogen has a positive effect on MAI of poplar plantations with denser stands and the efficiency of fertilization could be improved by matching the particular nutrient requirements of specific specie, as well as by altering planting densities. Using SEM, we attribute the variance of MAI to climatic and management factors (Fig. 2). The path coefficient is analogous to partial correlation coefficient, and describes the strength and sign of the relationships among the introduced variables (Grace, 2006). Our model also makes use of composite variables, which allow the effects of multiple conceptually related, but not necessarily statistically correlated, variables to be pooled into a single path to more closely match broader theoretical concepts (Grace and Bollen, 2008). In the final SEM, the path coefficient between the composite of climatic factors and MAI is positive, which is consistent with the results of bivariate models, however, the correlation between the composite of management factors and MAI displayed a negative relationship, suggesting current management
Fig. 3. Response of soil carbon sequestration and root carbon storage in poplar SRF to stand age. Solid lines indicate least-squares linear regression; Shaded area indicates confidence interval of 0.95. Equations and regression coefficients (R) were shown for significant relations.
D. Wang et al. / Environmental Research 144 (2016) 88–95
93
Fig. 4. Simulated carbon sequestration in short-rotation poplar forestry harvested at (a) 10-yr rotation cycles and (b) 5-yr rotation cycles in Inner Mongolia, China (Hu et al., 2008).
routines including harvesting cycle (stand age) and planting density might not be optimal for many trials. It is essential that potential confounding factors be considered in a meta-analysis which synthesizes results from a large number of studies that were conducted under a variety of climate and growing conditions. The climate data collected in this study were obtained from different resources and therefore represent the site climate with different accuracy and precision; a future improved analysis will require including more accurate climate data when available. SOC of poplar plantation is also quadraticly related to stand age, but showing an opposite trend with MAI and root carbon (Figs. 1 and 3). SOC sequestration is impacted by multiple factors including temperature, precipitation, soil and vegetation type, initial soil C stock and management practices (Sartori and Lal, 2006; Trumbore, 2006; Wang et al., 2015). Replanting usually results in significant losses of soil organic carbon (SOC), because inputs from new plants are too low to counteract losses by soil respiration, which is the case for the results of data analysis and model simulations (Figs. 3 and 4). Bashkin and Binkley (1998) found that site preparation caused an initial SOC loss and net recovery occurs about 10 years after plantation establishment. This may explain why short-term experiments usually fail to detect changes in SOC pools under poplar plantations (Coleman et al., 2004; Sartori et al., 2007). Our results are consistent with several studies that report long-term positive SOC changes following poplar afforestation of formerly cultivated lands (Hansen, 1993; Makeschin, 1994; Coleman et al., 2004). The dynamic simulations of poplar growth and SOC sequestration clearly demonstrate differences among the two harvesting scenarios in their capacity to produce biomass and sequestrate carbon (Fig. 4). The rate of SOC accumulation was about 0.8 Mg/ha/ yr with 10-yr rotation cycle, which was very close to what has been reported in a direct investigation wherein SOC accumulates 0.6 Mg/ha/yr (Pellegrino et al., 2011). 10-yr rotation harvest results in greater C sequestration relative to 5-yr rotation harvests, which is mainly caused by the build-up of SOC (Fig. 4). Plantation management results in gains or losses of C to the atmosphere depending on the harvest rotation interval and the duration of the practice. Short rotation forestry is a net source of C to the atmosphere in the short-term but this practice becomes a net sink of C over time with 10-yr harvest intervals (Fig. 4). The initial soil disturbance for planting results in a release of C because inputs from new plants are too low to counteract C losses by soil
respiration, but long-term growth eventually exceeds respiratory losses and makes SRF a net sink of C. For SRF with shorter rotation term, the growth of above-ground biomass and the build-up of SOC could not compensate for the C loss from the soil due to frequent distance of the soil. Though many studies have recommended the rotation length should last 4–6 years for poplars for optimal growth rate (Ceulemans and Deraedt, 1999; Liesebach et al., 1999; Mitchell et al., 1999), it might not be long enough for the soil to recover from the carbon loss from the initial establishment (Richter et al., 1999). Though commercial pressures dictate frequent harvests and short rotations, foresters should optimize the timing of harvesting to promote the most resilient species and carbon storage (Fares et al., 2015).
5. Conclusion The meta-analysis in this study indicates that poplar growth varies among different species/clones and across sites with different climate and management practice. There is an optimal range of Tmin, Tave and Tmax for Poplar growth, though accumulated growing season temperatures (GDD) has a linear positive effect with MAI. Poplar growth depends on water availability, as annual and seasonal precipitation both has a positive effect on poplar MAI. Poplar MAI reaches its maximum at the age of 17 and the effect of stand age on MAI covariates with planting density, with highest MAI-achieved stand age being earlier with denser stands. Nitrogen in this analysis displays no significant relationship with MAI, but at higher planting densities, nitrogen had a positive effect on MAI. The low production for poplar globally was probably caused by the suboptimal water availability, rotation length and planting density. The structural equation model attributes the variance of poplar growth more to climate than to management composite factors. The meta-analysis reveals that SOC in poplar plantations drops to lowest value with stand age increases and then started to increase at the stand age of 14 years. The case studies results from the Ecosystem Demography model further approves that harvesting at earlier stage might make the poplar plantation a net carbon source in a long term, suggesting that management factors of harvesting routine alone could have dramatic effect not only on the above ground growth but also on the soil carbon sequestration of poplar plantations. The findings of this study are useful in
94
D. Wang et al. / Environmental Research 144 (2016) 88–95
guiding further research, policy and management decisions towards sustainable poplar plantations.
Acknowledgments We appreciate the anonymous reviewers for giving their advice for the manuscript. Funding for this research was provided by Nanjing University of Information Science and Technology, the Jiangsu Natural Science Foundation (BK20150894), National Natural Science Foundation of China (31500503) and the International S&T Cooperation Program of China (2012DFA60830) through Dan Wang, the National Natural Science Foundation of China (61402236 and 61373064) through Yong Cheng and 973 Program (2012CB416904) through Honghua Ruan.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.envres.2015.10. 016.
References Anderson-Teixeira, K.J., Davis, S.C., Masters, M.D., DeLucia, E.H., 2009. Changes in soil organic carbon under biofuel crops. Glob. Chang. Biol. Bioenergy 1, 75–96. Arora, V.K., Boer, G.J., 2005. A parameterization of leaf phenology for the terrestrial ecosystem component of climate models. Glob. Chang. Biol. 11, 39–59. Auclair, D., Bouvarel, L., 1992. Influence of spacing and short rotations on Populus trichocarpa deltoids coppice. Can. J. Res. 22, 541–548. Aylott, M.J., Casella, E., Tubby, I., Street, N.R., Smith, P., Taylor, G., 2008. Yield and spatial supply of bioenergy poplar and willow short-rotation coppice in the UK. New Phytol. 178, 258–270. Bashkin, M.A., Binkley, D., 1998. Changes in soil carbon following afforestation in Hawaii. Ecology 79, 828–833. Baum, C., Leinweber, P., Weih, M., Lamersdorf, N., Dimitriou, L., 2009. Effects of short rotation coppice with willows and poplar on soil ecology. Agric. For. Res. 3, 183–196. Bellassen, V., Luyssaert, S., 2014. Managing forests in uncertain times. Nature 506, 153–155. Bollen, K.A., 1989. Structural Equations with Latent Variables. John Wiley & Sons, New York. Calfapietra, C., Gielen, B., Karnosky, D., Ceulemans, R., Scarascia Mugnozza, G., 2010. Response and potential of agroforestry crops under global change. Environ. Pollut. 158, 1095–1104. Canadian Council of Forest Ministers, 2001. Forest 2020 – a Dialogue with Canadians. Canadian Council of Forest Ministers Report. Available at 〈http://www. ccfm.org/forest2020/index_e.html〉. Ceulemans, R., Deraedt, W., 1999. Production physiology and growth potential of poplars under short-rotation forestry culture. For. Ecol. Manag. 121, 9–23. Coleman, M.D., Isebrands, J.G., Tolsted, D.N., Tolbert, V.R., 2004. Comparing soil carbon of short rotation Poplar plantations with agricultural crops and woodlots in North Central United States. Environ. Manag. 33, 299–308. Copenheaver, C.A., Crawford, C.J., Fearer, T.M., 2011. Age-specific responses to climate identified in the growth of Quercus alba. Trees 25, 647–653. D’Amato, A.W., Bradford, J.B., Fraver, S., Palik, B.J., 2013. Effects of thinning on drought vulnerability and climate response in north temperate forest ecosystems. Ecol. Appl. 23, 1735–1742. Davis, S.C., Dietze, M., DeLucia, E., Field, C., Hamburg, S.P., Loarie, S., Long, S.P., 2012. Harvesting carbon from eastern US forests: opportunities and impacts of an expanding bioenergy industry. Forests 3, 370–397. DeBell, D.S., Harrington, C.A., 1997. Productivity of Populus in monoclonal and polyclonal blocks at three spacings. Can. J. For. Res. 27, 978–985. Don, A., Osborne, B., Hastings, A., Skika, U., Carter, M.S., Drewer, J., Zenone, T., 2011. Land-use change to bioenergy production in Europe: implications for the greenhouse gas balance and soil carbon. Glob. Chang. Biol. Bioenergy 4, 372–391. Fares, S., Mugnozza, G.S., Corona, P., Palahi, M., 2015. Five steps for managing Europe’s forests. Nature 519, 407–409. Farquhar, G., Caemmerer, S., Berry, J., 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C 3 species. Planta 149, 78–90. Fu, Y.H., Campioli, M., Van Oijen, M., Deckmyn, G., Janssens, I.A., 2012. Bayesian comparison of six different temperature-based budburst models for four temperate tree species. Ecol. Model. 230, 92–100. Grace, J.B., 2006. Structural Equation Modeling and Natural Systems. Cambridge
University Press, Cambridge, Massachusetts, USA. Grace, J.B., Bollen, K.A., 2008. Representing general theoretical concepts in structural equation models: the role of composite variables. Environ. Ecol. Stat. 15, 191–213. Hansen, E.A., 1993. Soil carbon sequestration beneath hybrid poplar plantations in the north central United States. Biomass Bioenerg. 5, 431–436. Heilman, P.E., Xie, F.G., 1993. Influence of nitrogen on growth and productivity of short-rotation Populus trichocarpa-Populus deltoides hybrids. Can. J. Res. 23, 1863–1869. Hill, J., Nelson, E., Tilman, D., Polasky, S., Tiffany, D., 2006. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. USA 103, 11206–11210. Hu, Y.L., Zheng, D.H., Fan, Z.P., Chen, G.S., Zhao, Q., Pepper, D., 2008. Changes in ecosystem carbon stocks following grassland afforestation of semiarid sandy soil in the southeastern Keerqin Sandy Lands, China. J. Arid Environ. 72 (12), 2193–2200. Kauter, D., Lewandowski, I., Claupein, W., 2003. Quantity and quality of harvestable biomass from Populus short rotation coppice for solid fuel use – a review of the physiological basis and management influences. Biomass Bioenergy 24, 411–427. Kline, R.B., 2005. Principles and Practice of Structural Equation Modeling, Second edition. Guilford Press, New York, USA. Kolström, M., Lindner, M., Vilén, T., Maroschek, M., Seidl, R., Lexer, M.J., et al., 2011. Reviewing the science and implementation of climate change adaptation measures in European forestry. Forests 2, 961–982. http://dx.doi.org/10.3390/ f2040961. Leonelli, G., Denneler, B., Bergeron, Y., 2008. Climate sensitivity of trembling aspen radial growth along a productivity gradient in northeastern British Columbia, Canada. Can. J. For. Res. 38, 1211–1222. Leuning, R., 1995. A critical appraisal of a combined stomatal photosynthesis model for C3 plants. Plant Cell Environ. 18 (4), 339–355. Liesebach, M., Wuehlisch, G.V, Muhs, H.-J., 1999. Aspen for short-rotation coppice plantations on agricultural sites in Germany: Effects of spacing and rotation length on growth and biomass production of aspen progenies. For. Ecol. Manag. 121, 25–39. Linares, J.C., Camarero, J.J., Carreira, J.A., 2010. Competition modulates the adaptation capacity of forests to climatic stress: insights from recent growth decline and death in relic stands of the Mediterranean fir Abies pinsapo. J. Ecol. 98, 592–603. Makeschin, F., 1994. Effects of energy forestry on soils. Biomass Bioenergy 6, 63–69. McLane, S.C., Daniels, L.D., Aitken, S.N., 2011. Climate impacts on lodgepole pine (Pinus contorta) radial growth in a provenance experiment. For. Ecol. Manag. 262, 115–123. Mead, D.G., 2005. Opportunities for improving plantation productivity. How much? How quickly? How realistic?. Biomass Bioenergy 28, 249–266. Meiresonne, L., Schrijver, A.D., Vos, B.D., 2006. Nutrient cycling in a poplar plantation (Populus trichocarpa Populus deltoides, Beaupre´) on former arable land in northern Belgium. Can. J. Res. 37, 141–155. Mérian, P., Lebourgeois, F., 2011. Size-mediated climate-growth relationships in temperate forests: a multi-species analysis. For. Ecol. Manag. 261, 1382–1391. Mitchell, C.P., Stevens, E.A., Watters, M.P., 1999. Short rotation forestry operations, productivity and costs based on experience gained in the UK. Ecol. Manag. 121, 123–136. Moorcroft, P.R., Hurtt, G.C., Pacala, S.W., 2001. A method for scaling vegetation dynamics: the ecosystem demography model (ED). Ecol. Monogr. 74, 557–586. Orwig, D.A., Abrams, M.D., 1997. Variation in radial growth responses to drought among species, site, and canopy strata. Trees 11, 474–484. Pan, C., Tajchman, S.J., Kochenderfer, J.N., 1997. Dendroclimatological analysis of major forest species of the central Appalachians. For. Ecol. Manag. 98, 77–87. Parton, W., Scurlock, J., Ojima, D., Scholes, R., Schimel, D., Gilmanov, T., Kirchner, T., Menaut, J.C., Seastedt, T., Moya, E.G., Kamnalrut, A., Kinyamario, J., 1993. Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide. Glob. Biogeochem. Cycles 7, 785–809. Pellegrino, E., Di Bene, C., Tozzini, C., Bonari, E., 2011. Impact on soil quality of a 10year-old short-rotation coppice poplar stand compared with intensive agricultural and uncultivated systems in a Mediterranean area. Agric. Ecosyst. Environ. 140, 245–254. Richter, D.D., Markewitz, D., Trumbore, S.E., Wells, C.G., 1999. Rapid accumulation and turnover of soil carbon in a re-establishing forest. Nature 400, 56–58. Saldarriaga, J.G., West, D.C., Tharp, M.L., Uhl, C., 1988. Long-term chronosequence of forest succession in the up-per Rio Negro of Colombia and Venezuela. J. Ecol. 76, 938–958. Sartori, F., Lal, R., 2006. Potential soil carbon sequestration and CO2 offset by dedicated energy crops in the USA. Crit. Rev. Plant Sci. 25, 441–472. Sartori, F., Markewitz, D., Borders, B., 2007. Soil carbon storage and nitrogen and phosphorus availability in loblolly pine plantations over 4 to 16 years of herbicide and fertilizer treatments. Biochemistry 84, 13–30. Sellers, P.J., 1992. Biophysical models of land processes. In: Trenberth, K.E. (Ed.), Climate System Modeling. Cambridge University Press, Cambridge, UK, pp. 451–490. Somerville, C., Young, H., Taylor, C., Davis, S.C., Long, S.P., 2010. Feedstocks for lignocellulosic biofuels. Science 329, 790. Stanturf, J.A., Oosten, C., Netzer, D.A., Coleman, M.D., Portwood, C.J., 2001. Ecology and silviculture of poplar plantations. In: Dickmann, D.I., Isebrands, J.G., Eckenwalder, J.E., Richardson, J. (Eds.), Poplar Culture in North America. NRC Research Press, Ottawa, pp. 153–206.
D. Wang et al. / Environmental Research 144 (2016) 88–95
Trumbore, S., 2006. Carbon respired by terrestrial ecosystems-recent progress and challenges. Glob. Chang. Biol. 12, 141–153. Tullus, A., Tullus, H., Vares, A., Kanal, A., 2007. Early growth of hybrid aspen (Populus wettsteinii Hä met-Ahti) plantations on former agricultural lands in Estonia. For. Ecol. Manag. 245, 118–129. Ulzen-Appiah, F., Briggs, R.D., Abrahamson, L.P., Bickelhaupt, D.H., 2000. Soil carbon pools in short-rotation willow (Salix dasyclados) plantation four years after establishment. In: Proceedings of Bioenergy 2000, Buffalo, NY, October 15–19. Van de Walle., I., Van Camp, N., Van de Casteele, L., Verheyen, K., Lemeur, R., 2007. Short-rotation forestry of birch, maple, poplar and willow in Flanders (Belgium) I. Biomass production after 4 years of tree growth. Biomass Bioenergy 31, 267–275. Viovy, N., Ciais, P., 2009. A Combined Dataset for Eecosystem Modelling. 〈http:// dods.extra.cea.fr/10 data/p529viov/cruncep/readme.htm〉. Walko, R.L., Band, L.E., Baron, J., et al., 2000. Coupled atmosphere–biophysics–hydrology models for environmental modeling. J. Appl. Meteorol. 39, 931–944.
95
Wang, D., LeBauer, D.S., Dietze, M.C., 2010. A quantitative review comparing the yield of switchgrass in monocultures and mixtures in relation to climate and management factors. Glob. Chang. Biol. Bioenergy 2, 16–25. Wang, D., LeBauer, D., Kling, G., Voigt, T., Dietze, M.C., 2013. Ecophysiological screening of tree species for biomass production: trade-off between production and water use. Ecosphere 4 (11), 138. Wang, D., LeBauer, D.S., Dietze, M.C., 2012. Predicting yields of short-rotation hybrid poplar (Populus spp.) for the United States through model–data synthesis. Ecol. Appl. 23, 944–958. Wang, D., Jaiswal, D., Lebauer, D.S., Wertin, T.M., Bollero, G.A., Leakey, A.D., Long, S. P., 2015. A physiological and biophysical model of coppice willow (Salix spp.) production yields for the contiguous USA in current and future climate scenarios. Plant Cell Environ. 38, 1850–1865. http://dx.doi.org/10.1111/pce.12556. Weih, M., 2004. Intensive short rotation forestry in boreal climates: present and future perspective. Can. J. For. Res. 34, 1369–1378.
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
Analyzing the impact of climate and management factors on the productivity and soil carbon sequestration of poplar plantations Dan Wang a,n, Jiazhi Fan a, Panpan Jing a, Yong Cheng a, Honghua Ruan b a International Center for Ecology, Meteorology and Environment, School of Applied Meteorology, Nanjing University of Information Science and Technology, Nanjing 210044, PR China b Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 9 May 2015 Received in revised form 9 October 2015 Accepted 15 October 2015 Available online 31 October 2015
It is crucial to investigate how climate and management factors impact poplar plantation production and soil carbon sequestration interactively. We extracted above-ground net primary production (ANPP), climate and management factors from peer-reviewed journal articles and analyzed impact of management factor and climate on the mean annual increment (MAI) of poplar ANPP statistically. Previously validated mechanistic model (ED) is used to perform case simulations for managed poplar plantations under different harvesting rotations. The meta-analysis indicate that the dry matter MAI was 6.3 Mg ha 1 yr 1 (n ¼ 641, sd ¼4.9) globally, and 5.1 (n ¼292, sd¼ 4.0), 8.1 (n ¼ 224, sd¼ 4.7) and 4.4 Mg ha 1 yr 1 (n ¼125, sd ¼3.2) in Europe, the US and China, respectively. Poplar MAI showed a significant response to GDD, precipitation and planting density and formed a quadratic relationship with stand age. The low annual production for poplar globally was probably caused by suboptimal water availability, rotation length and planting density. SEM attributes the variance of poplar growth rate more to climate than to management effects. Case simulations indicated that longer rotation cycle significantly increased soil carbon storage. Findings of this work suggests that management factor of rotation cycle alone could have dramatic impact on the above ground growth, as well as on the soil carbon sequestration of poplar plantations and will be helpful to quantify the long-term carbon sequestration through short rotation plantation. The findings of this study are useful in guiding further research, policy and management decisions towards sustainable poplar plantations. & 2015 Elsevier Inc. All rights reserved.
Keywords: Meta-analysis Short rotation forestry Structural equation model Ecosystem Demography model Stand age Planting density
1. Introduction Hybrid poplars are among the most widely cultivated hardwood species for pulp and timber production due to their fast growth rate, high light-use efficiency and photosynthetic capacity, strong tolerance to biophysical stress, ease of vegetative propagation and adaptation to a wide variety of soils (Weih, 2004). There is recognition that fast-growing, high-yield plantations will be important in meeting an increasing global demand for wood and biofuel products to substitute fossil fuels and carbon intensive materials such as steel (Canadian Council of Forest Ministers, 2001; Somerville et al., 2010). Understanding how poplar tree Abbreviations: ANPP, above-ground net primary production; ED, Ecosystem Demography model; GDD, growing degree days; MAI, mean annual increment of ANPP; Pa, annual precipitation; Ps, growing season precipitation; SEM, structural equation model; SOC, soil organic carbon; SRF (C), short rotation forestry (coppice); Tave, annual average temperature; Tmax, annual maximum temperature; Tmin, annual minimum temperature n Corresponding author. E-mail address: [email protected] (D. Wang). http://dx.doi.org/10.1016/j.envres.2015.10.016 0013-9351/& 2015 Elsevier Inc. All rights reserved.
grows has become critically important in recent years as the climate warms and estimates of biomass of forest products (including bioenergy products) and stored forest carbon (50 per cent of forest biomass) are needed. Poplar growth rate vary considerably and are affected by the influence of genetics, climate, and management factors on survival, competition, and vigor of the stand (Mead, 2005). The specific response of poplar growth to climate varies across Populus species (Pan et al., 1997), tree size (Mérian and Lebourgeois, 2011), age (Copenheaver et al., 2011), stand structures (Linares et al., 2010; D'Amato et al., 2013), edaphic or productivity gradients (Orwig and Abrams, 1997; Leonelli et al., 2008), and genetic variability across populations (McLane et al., 2011). Although climate influences tree growth, other management practice such as nitrogen fertilization, harvesting practice and planting density can interact with climate to further affect tree growth (Davis et al., 2012). Past studies linking tree growth with climate have often failed to consider the interacting effects of stand characters and management practices. This likely over-simplifies climate-growth relationships and the potential effect of climate and management practice on both tree- and stand-level productivity. Separating the
D. Wang et al. / Environmental Research 144 (2016) 88–95
impact of climate and management factors on tree growth is useful to guide plantation management in adapting to future climate change conditions. The cultivation of fast-growing woody plants within short rotation forestry (SRF) can provide a potentially important source of wood production, environmental protection and economy income. Woody species managed by short rotation forestry (SRF) have multiple advantages over annual crops which include enhanced biological diversity, greater carbon sequestration, and reduced inputs of labor, pesticides and fertilizers (Hill et al., 2006; Baum et al., 2009; Don et al., 2011). Rotations shorter than 3 years, however, could lead to reduced yields after several rotations due to physiological problems including stump aging and depletion of carbohydrate reserves (Auclair and Bouvarel, 1992) and causes the depletion of nitrogen due to greater consumption of nitrogen than the supply. The maximum biomass productivity is expected with harvest cycles of 3–11 years (Sartori and Lal, 2006). As poplar increased in age, there was a major reduction in nutrient concentrations in the leaves, indicating that nitrogen limitation might become more evident when poplar stand gets older (Wang et al., 2013). Although aboveground woody biomass is the economically important component of SRF ecosystems, enhanced carbon sequestration in roots and soil has a large impact on CO2 mitigation and the ecological benefits of SRF related to conservation, water and soil protection, recreation or climate-change mitigation and adaptation are likely to acquire economic value in the future (Calfapietra et al., 2010). SRF can rapidly accumulate C in stable components such as stems, branches and coarse roots, while at the same time cycling C and nutrients to the soil through more labile litter pools consisting of leaves, twigs and fine roots (Sartori and Lal, 2006; Meiresonne et al., 2006). Observed patterns of soil organic carbon (SOC) dynamics under SRF include short-term losses (Hansen, 1993), long term gains (Hansen, 1993; Makeschin, 1994) and no changes (Ulzen-Appiah et al., 2000). Therefore, determining the optimal rotation cycle will not only be beneficial to poplar yields but also the carbon sequestration level. While the yield potential of poplar has been tested by many on-farm studies and has been modeled and reviewed extensively, to date there have been no quantitative reviews on how climate and management practice affect poplar productivity interactively. And information is scarcer on how rotation cycles affect carbon sequestration in the soil (Anderson-Teixeira et al., 2009). Structural equation modeling is a scientific methodology that aspires to make a strong and explicit connection between empirical data and theoretical ideas (Bollen, 1989; Kline, 2005). It can partition causal influences among multiple variables, allowing the separation effects of different predictors (Grace, 2006). In the present study, we will analyze the effect of climate and management factors on the mean annual increment of poplar ANPP (MAI) intensively and provide guidelines for plantation management to optimize poplar yields and soil carbon sequestration in SRF based on a new, updated quantitative literature review. A structural equation model will be applied to separate the partial effect of climate and management factors on poplar ANPP. We will further run a well-calibrated model on a case site to test the effect of two different rotation cycles on carbon sequestration in poplar SRF. Sustainable forest management practices could maximize carbon sequestration and maintain stand quality and productivity; therefore, the primary goal of this study was to examine the effect of climate and management practice on the growth of poplar across broad geographical ranges. Such quantitative information out of meta-analysis and mechanical models will be used to (1) understand the productivity variability of poplar plantations under different circumstances; (2) identify the key factors impacting poplar MAI and soil carbon sequestration and separating different impacts of climate and management factors on poplar
89
productivity; (3) evaluating the effect of different harvesting cycles on SRF carbon sequestration.
2. Methods The peer-reviewed journal articles used to construct the database for this meta-analysis were obtained by searching the Science Citation Index (SCI) of the Institute of Scientific Information and Chinese Journal Full-text Database (CJFD). The inclusion of journal articles in Chinese ensured the completeness of the growth information for poplar plantation, as poplar plantation area in China is not negligible but journal articles published in Chinese often being excluded in meta-analysis due to languages barriers. The list of articles obtained was subsequently cross-checked with references cited in a large number of review articles and books to ensure the inclusion of all articles containing data relevant for this meta-analysis. Any article published before the end of 2014 that includes the following information was included: (1) biomass of poplar plantation; (2) site location; (3) management information on planting density, stand age, nitrogen fertilization level. Trials with pathogen or disease attacks were excluded from the analysis. In total, 29 peer-reviewed articles from China, 28 from other countries were included in this meta-analysis (Appendix A and B). Meteorological data was obtained from published articles or nearby meteorological stations if available or from LOCCLIM (v. 1.0 FAO, Rome, Italy) for a given site when climate data were not available (Wang et al., 2010). Growing degree days (GDD) were calculated in the growing season with the base temperature of 10 °C. The growing season is defined from last frost in spring to the first frost season in autumn or the date of harvest. We assumed studies conducted at different sites, yields from different treatments (e.g. fertilizer treatments), and different growing seasons were independent. In our analysis, studies in which plants were grown under environmental stresses (e.g., drought, low nutrients, light deficiency and etc.) were excluded. The database of poplar plantation contains climate information including annual average (Tave), maximum (Tmax) and minimum temperature (Tmin), GDD, growing season precipitation (Ps), annual precipitation (Pa), site information (site location, longitude, latitude, altitude, soil type), management routine (planting date, rotation scheme, nitrogen level, planting density, stand age, herbicide), biomass and soil C. For studies with more detailed growth information, carbon storage in leaves, branches, trunk, roots and litter are also included in the database. Graphical data were extracted from the articles using digitizing software (GET DATA GRAPH DIGITIZER v. 2.22). Data were firstly sorted and tested for normality. Squared-root transformed MAI were then analyzed using mixed model analysis of variance. The random effects in the mixed model framework were used to account for site-to-site variability that were not accounted for by the fixed-effect covariates but which potentially caused treatments within sites to not be independent (e.g. soil types, soil micro fauna and etc.). Similarly, species random effects account for the differences that could not be accounted for as fixed effects due to limited and unbalanced replication data (Wang et al., 2010). Stepwise regression was conducted to screen a best model to describe MAI (only two way interactions were considered in the analysis). A mixed model taking individual studies and species as random effects was conducted to test the fixed effects of annual average (Tave), maximum (Tmax) and minimum temperature (Tmin), GDD, growing season precipitation (Ps), annual precipitation (Pa), stand age, planting density, N and their interactions on MAI. Prior to evaluating multivariate models, bivariate relations between all variables in the model were assessed. We examined scatterplots for the presence of outliers, evidence of skewness or kurtosis, and non-linear relationships up to second-order
90
D. Wang et al. / Environmental Research 144 (2016) 88–95
polynomials. The relationship between poplar MAI and SOC and individual climate and management factors was evaluated separately by linear or second-order polynomial regression analysis. Tests of hypotheses were considered significant at P r0.05 level. As a second multivariate approach to examine the relationship of poplar productivity with climate and management factors proportionally, structural equation modeling (SEM) was conducted to separate the effect of climate and management. Maximum likelihood solution procedures and chi-square goodness of fit measures were employed to evaluate model adequacy. Finally, we used a well calibrated and validated model, Ecosystem Demography model (ED2), to simulate the productivity and soil carbon sequestration of short-rotation cropping systems of hybrid poplar (Moorcroft et al., 2001; Wang et al., 2012). Established models of leaf level physiology (Farquhar et al., 1980; Leuning, 1995), allocation (Saldarriaga et al., 1988), biogeochemistry (Parton et al., 1993), land surface biophysics (Walko et al., 2000), and hydrology (Sellers, 1992) are used to simulate plant growth in ED2. Model simulations for poplar SRF were run for 68 years based on a 5-yr or 10-yr rotation cycles at a trial site at Inner Mongolia, China (Hu et al., 2008), with the same parameter sets as used in Wang et al. (2012). ED2 simulates biogeochemical as well as growing processes, which enables ED2 not only predict poplar growth but also the carbon cycling in the soil. The simulation assumed no manipulations of water, N, or P. Meteorological drivers were obtained from cru-NCEP reanalysis data and were sequentially cycled over years from 1981 to 2008 (Viovy and Ciais, 2009). Soil depth and textural class for each site were obtained from the published literature. Forest harvesting in ED2 is treated as a removal of the above ground stem biomass, with roots and leaves entering the litter pool.
3. Results The mean dry matter MAI of poplar in China is 4.4 Mg ha 1 yr 1 (n ¼125, sd ¼3.2) and significantly lower than that in European (5.1 Mg ha 1 yr 1, n ¼ 292, sd ¼4.0) and American countries (8.1 Mg ha 1 yr 1, n ¼224, sd¼ 4.7), and even the average globally (6.3 Mg ha 1 yr 1, n ¼641, sd ¼4.9) (Table 1). For MAI, the variance components of the random effect of individual studies and species were significantly different from 0, suggesting that individual studies and species did differ in their MAI (Table 2). The stepwise regression analysis indicated the set of five factors including Ps, GDD, stand age, planting density, N and some of their interactions significantly affected MAI and explained MAI the best (Table 1). The bivariate models indicate that GDD, Ps and Pa, and planting density had a linear relationship with MAI, and Tmin, Tave, Tmax and stand age formed a quadratic relationship with MAI, while N had no relationship with MAI (Fig 1). In order to separate the effect of climate and management factors on MAI, we conducted a SEM on the observed data. The final SEM fits the data well (X2 ¼12.3, d.f. ¼3, P¼ 0.16). The model indicates that climate and management composite factors significantly influenced MAI. All explanatory variables were significantly correlated with each other, indicating that they were not completely independent. The climate composite factors had a positive correlation and management composite factors a negative correlation with MAI, with path coefficient of 0.63 and 0.32, respectively (Fig. 2). Soil carbon sequestration formed a quadratic relationship with stand age, reaching the lowest value at about 14-yr (Fig. 3A). Root carbon storage, however, reached the highest value at 17-yr (Fig. 3B). We simulated annual wood production and net C sequestration of poplar plantations in a case study located at Inner Mongolia,
Table 1 The mean MAI of poplar plantations in China, Europe, the United State. Regions
MAI
n
sd
Global Europe North America China
6.3 5.1 8.1 4.4
641 292 224 125
4.9 4.0 4.7 3.2
Table 2 Fixed effects from mixed model of stand age, growing degree days (GDD), seasonal precipitation (Ps), planting density, N and their interactions on the mean maximum annual increment of aboveground production, taking individual studies and species as random effects. Effects
F-value
Pr4F
Stand age Planting density N GDD Ps Stand age Planting density Stand age N Stand age Ps Planting density GDD GDD N GDD Ps Planting density N
23.36 5.14 6.96 45.97 41.37 35.74 2.17 0.15 42.85 2.99 8.54 8.59
o 0.001 o 0.05 o 0.001 o 0.001 o 0.001 o 0.001
o 0.001 o 0.001 o 0.001
For all the effects, the dfN (numerator degrees of freedom)¼ 1 and the dfD (denominator degrees of freedom)¼ 643 For the random effect of individual studies: Z¼32.4, P o 0.05. For the random effect of species: Z¼44.6, Po 0.05.
that were completely harvested at either 10- or 5-yr intervals using the ED model (Hu et al., 2008; Wang et al., 2012). The MAI in the first harvesting cycle was comparable to the observed (data not shown). If grown in 10-yr cycles, the model simulations of net C storage (including aboveground biomass and soil carbon accumulation) were negative for the first 22 years and recovery occurred after about 3 harvesting cycles (Fig. 4). With a 5-yr harvesting cycle, the net C storage continued to decline over the 68-yr period (Fig. 4). In the longer harvesting cycle, the poplar plantation was a net sink of C after about 30 years, but in the shorter harvesting cycle, the poplar plantation was a net source of C to the atmosphere for the entire period of time tested.
4. Discussion The global average mean MAI of poplar is lower than the reported harvestable yields of poplar in temperate regions of Europe and North America between 10 and 15 Mg ha 1 yr 1 (Kauter et al., 2003; Van de Walle. et al., 2007; Aylott et al., 2008). This is especially true for the MAI in China, where it is significantly lower than that in Europe, USA and even the global average. Poplar plantation in China has reached more than 700 million hectares and ranked the largest in the world, exceeding the total area of poplar plantation in other countries. Therefore, future research understanding why the poplar MAI in China is not comparable to the global mean is urgently needed. The growth rate of poplar varies among different species/clones and sites and depends on the climate and management factors and their interactions (Table 2 and Fig. 1). The set of five factors including Pr, GDD, stand age, planting density, N and the interactions between stand age and planting density, between GDD and planting density, between GDD and Pr, and between planting density and N significantly affects MAI and explained MAI the best. Temperature factors, including Tmin, Tave and Tmax, all have a
D. Wang et al. / Environmental Research 144 (2016) 88–95
91
Fig. 1. Response of poplar MAI to annual average (Tave), maximum (Tmax) and minimum (Tmin) temperature, GDD, growing season precipitation (Ps), annual precipitation (Pa), stand age, planting density and N. Solid lines indicate least-squares linear regression. Shaded area indicates confidence interval of 0.95. Equations and regression coefficients (R) were shown for significant relations.
significant quadratic relationship with poplar MAI, suggesting that there is an optimal temperature range for poplar growth. GDD and precipitation both affect poplar productivity significantly and have a positive relationship with poplar MAI. Seasonal and annual precipitation has a similar impact on poplar MAI, which is consistent with many previous studies that approved the dependence of poplar growth on water availability (Davis et al., 2012; Wang et al., 2013). Among the studied soil properties, soil texture and moisture condition had the strongest and most reliable influence on mean growth characteristics in 5-year-old hybrid aspen plantations (Tullus et al., 2007). The same tendency has been observed also in North America for poplars in general (Stanturf et al., 2001). GDD can affect poplar MAI through affecting the phenology of flowering time and nutrient translocation (Arora and Boer, 2005),
which in turn will affect the annual aboveground productivity. Fu et al. (2012) found that higher GDD will advance bud burst in spring, put off leaf fall in autumn and extend the growing season. The fact that species/clone choice and climatic factors affects poplar ANPP significantly indicates that managers should select species that tolerate a variety of climates, such as those that can grow over a range of latitudes and altitudes (Kolström et al., 2011; Bellassen and Luyssaert, 2014; Fares et al., 2015). Stand age that maximizes the MAI can be used to inform the time period of rotation cycles. SRF includes single stem production followed either by replanting or by coppicing (Weih, 2004). The studies included in this analysis contain both replanting and coppicing trials; therefore results would vary and in general coppicing will enable a shorter rotation cycle and produce higher
92
D. Wang et al. / Environmental Research 144 (2016) 88–95
Fig. 2. Final structural equation model results (X2 ¼12.3, d.f. ¼ 3, P ¼0.16) with standardized path coefficients. The theoretical construct is represented by latent variables (circle) and all measured variables (rectangles) are represented as effect indicators. The proportion of variance explained (R2) is given beside the response variables in the model. Error terms represent unexplained variance of dependent variables. All pathways were significant (Po 0.05). The goodness of fit for the models was determined using a χ2-test.
yields. Nevertheless, we still found that stand age has a quadratic relationship with MAI, indicating at a stand age of about 17-yr, the MAI reached its maximum (Fig. 1G). The MAI-achieving year calculated here should be integrated with the knowledge on morphological and physiological factors influencing the regeneration and sustainability of the plantation, as well as the economic costs of harvesting and post-harvest processes to evaluate the relative benefits of different rotation cycles (Wang et al., 2013). Furthermore, the current MAI of polar might reach its maximum earlier than the mean MAI. For example, the maximum annual increment of ANPP was reached at the age of 13 years and the MAI at the age of 24 years (Liesebach et al., 1999). The stand age effect on MAI depends on the planting density, and with higher-density stands, MAI reaches maximum in an earlier stage. Planting density had a negative linear relationship with MAI,
mainly caused by the low MAI achieved at very high planting densities, which is consistent with previous studies (DeBell and Harrington, 1997) (Fig. 1H). Ceulemans and Deraedt (1999) stated that for poplar SRF, optimal densities range from 2500 to 10,000 trees ha 1. Even though the canopy and leaf properties of poplar SRF are very dynamic and tend to maximize light interception by the canopy, the fast growth and rapid changes in canopy depth and leaf area during the growing season require an optimal planting density with a dynamic and optimal distribution of foliage area, leaf mass and specific leaf area of the poplar canopy. Thinning can encourage new growth and increase structural diversity, which will increase canopy complexity and promote more efficient use of light and nitrogen (Fares et al., 2015). Many studies have illustrated the beneficial effect of nitrogen treatments on poplar growth and productivity. But due to large differences in fertilization response and maximum growth observed among various hybrid crosses (Heilman and Xie, 1993; Ceulemans and Deraedt, 1999), we could not find a positive effect of N on MAI in this analysis. And contrary to our expectation that N demand might increase as stands age due to N removal from biomass harvesting and leaching out of root zone, the response of poplar MAI to N x stand age treatment is not significant (Table 1). Makeschin and Rehfuess (1994) found no effect in biomass production in fertilized and unfertilized trials. There is a significant N x planting density effect on MAI, suggesting that nitrogen has a positive effect on MAI of poplar plantations with denser stands and the efficiency of fertilization could be improved by matching the particular nutrient requirements of specific specie, as well as by altering planting densities. Using SEM, we attribute the variance of MAI to climatic and management factors (Fig. 2). The path coefficient is analogous to partial correlation coefficient, and describes the strength and sign of the relationships among the introduced variables (Grace, 2006). Our model also makes use of composite variables, which allow the effects of multiple conceptually related, but not necessarily statistically correlated, variables to be pooled into a single path to more closely match broader theoretical concepts (Grace and Bollen, 2008). In the final SEM, the path coefficient between the composite of climatic factors and MAI is positive, which is consistent with the results of bivariate models, however, the correlation between the composite of management factors and MAI displayed a negative relationship, suggesting current management
Fig. 3. Response of soil carbon sequestration and root carbon storage in poplar SRF to stand age. Solid lines indicate least-squares linear regression; Shaded area indicates confidence interval of 0.95. Equations and regression coefficients (R) were shown for significant relations.
D. Wang et al. / Environmental Research 144 (2016) 88–95
93
Fig. 4. Simulated carbon sequestration in short-rotation poplar forestry harvested at (a) 10-yr rotation cycles and (b) 5-yr rotation cycles in Inner Mongolia, China (Hu et al., 2008).
routines including harvesting cycle (stand age) and planting density might not be optimal for many trials. It is essential that potential confounding factors be considered in a meta-analysis which synthesizes results from a large number of studies that were conducted under a variety of climate and growing conditions. The climate data collected in this study were obtained from different resources and therefore represent the site climate with different accuracy and precision; a future improved analysis will require including more accurate climate data when available. SOC of poplar plantation is also quadraticly related to stand age, but showing an opposite trend with MAI and root carbon (Figs. 1 and 3). SOC sequestration is impacted by multiple factors including temperature, precipitation, soil and vegetation type, initial soil C stock and management practices (Sartori and Lal, 2006; Trumbore, 2006; Wang et al., 2015). Replanting usually results in significant losses of soil organic carbon (SOC), because inputs from new plants are too low to counteract losses by soil respiration, which is the case for the results of data analysis and model simulations (Figs. 3 and 4). Bashkin and Binkley (1998) found that site preparation caused an initial SOC loss and net recovery occurs about 10 years after plantation establishment. This may explain why short-term experiments usually fail to detect changes in SOC pools under poplar plantations (Coleman et al., 2004; Sartori et al., 2007). Our results are consistent with several studies that report long-term positive SOC changes following poplar afforestation of formerly cultivated lands (Hansen, 1993; Makeschin, 1994; Coleman et al., 2004). The dynamic simulations of poplar growth and SOC sequestration clearly demonstrate differences among the two harvesting scenarios in their capacity to produce biomass and sequestrate carbon (Fig. 4). The rate of SOC accumulation was about 0.8 Mg/ha/ yr with 10-yr rotation cycle, which was very close to what has been reported in a direct investigation wherein SOC accumulates 0.6 Mg/ha/yr (Pellegrino et al., 2011). 10-yr rotation harvest results in greater C sequestration relative to 5-yr rotation harvests, which is mainly caused by the build-up of SOC (Fig. 4). Plantation management results in gains or losses of C to the atmosphere depending on the harvest rotation interval and the duration of the practice. Short rotation forestry is a net source of C to the atmosphere in the short-term but this practice becomes a net sink of C over time with 10-yr harvest intervals (Fig. 4). The initial soil disturbance for planting results in a release of C because inputs from new plants are too low to counteract C losses by soil
respiration, but long-term growth eventually exceeds respiratory losses and makes SRF a net sink of C. For SRF with shorter rotation term, the growth of above-ground biomass and the build-up of SOC could not compensate for the C loss from the soil due to frequent distance of the soil. Though many studies have recommended the rotation length should last 4–6 years for poplars for optimal growth rate (Ceulemans and Deraedt, 1999; Liesebach et al., 1999; Mitchell et al., 1999), it might not be long enough for the soil to recover from the carbon loss from the initial establishment (Richter et al., 1999). Though commercial pressures dictate frequent harvests and short rotations, foresters should optimize the timing of harvesting to promote the most resilient species and carbon storage (Fares et al., 2015).
5. Conclusion The meta-analysis in this study indicates that poplar growth varies among different species/clones and across sites with different climate and management practice. There is an optimal range of Tmin, Tave and Tmax for Poplar growth, though accumulated growing season temperatures (GDD) has a linear positive effect with MAI. Poplar growth depends on water availability, as annual and seasonal precipitation both has a positive effect on poplar MAI. Poplar MAI reaches its maximum at the age of 17 and the effect of stand age on MAI covariates with planting density, with highest MAI-achieved stand age being earlier with denser stands. Nitrogen in this analysis displays no significant relationship with MAI, but at higher planting densities, nitrogen had a positive effect on MAI. The low production for poplar globally was probably caused by the suboptimal water availability, rotation length and planting density. The structural equation model attributes the variance of poplar growth more to climate than to management composite factors. The meta-analysis reveals that SOC in poplar plantations drops to lowest value with stand age increases and then started to increase at the stand age of 14 years. The case studies results from the Ecosystem Demography model further approves that harvesting at earlier stage might make the poplar plantation a net carbon source in a long term, suggesting that management factors of harvesting routine alone could have dramatic effect not only on the above ground growth but also on the soil carbon sequestration of poplar plantations. The findings of this study are useful in
94
D. Wang et al. / Environmental Research 144 (2016) 88–95
guiding further research, policy and management decisions towards sustainable poplar plantations.
Acknowledgments We appreciate the anonymous reviewers for giving their advice for the manuscript. Funding for this research was provided by Nanjing University of Information Science and Technology, the Jiangsu Natural Science Foundation (BK20150894), National Natural Science Foundation of China (31500503) and the International S&T Cooperation Program of China (2012DFA60830) through Dan Wang, the National Natural Science Foundation of China (61402236 and 61373064) through Yong Cheng and 973 Program (2012CB416904) through Honghua Ruan.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.envres.2015.10. 016.
References Anderson-Teixeira, K.J., Davis, S.C., Masters, M.D., DeLucia, E.H., 2009. Changes in soil organic carbon under biofuel crops. Glob. Chang. Biol. Bioenergy 1, 75–96. Arora, V.K., Boer, G.J., 2005. A parameterization of leaf phenology for the terrestrial ecosystem component of climate models. Glob. Chang. Biol. 11, 39–59. Auclair, D., Bouvarel, L., 1992. Influence of spacing and short rotations on Populus trichocarpa deltoids coppice. Can. J. Res. 22, 541–548. Aylott, M.J., Casella, E., Tubby, I., Street, N.R., Smith, P., Taylor, G., 2008. Yield and spatial supply of bioenergy poplar and willow short-rotation coppice in the UK. New Phytol. 178, 258–270. Bashkin, M.A., Binkley, D., 1998. Changes in soil carbon following afforestation in Hawaii. Ecology 79, 828–833. Baum, C., Leinweber, P., Weih, M., Lamersdorf, N., Dimitriou, L., 2009. Effects of short rotation coppice with willows and poplar on soil ecology. Agric. For. Res. 3, 183–196. Bellassen, V., Luyssaert, S., 2014. Managing forests in uncertain times. Nature 506, 153–155. Bollen, K.A., 1989. Structural Equations with Latent Variables. John Wiley & Sons, New York. Calfapietra, C., Gielen, B., Karnosky, D., Ceulemans, R., Scarascia Mugnozza, G., 2010. Response and potential of agroforestry crops under global change. Environ. Pollut. 158, 1095–1104. Canadian Council of Forest Ministers, 2001. Forest 2020 – a Dialogue with Canadians. Canadian Council of Forest Ministers Report. Available at 〈http://www. ccfm.org/forest2020/index_e.html〉. Ceulemans, R., Deraedt, W., 1999. Production physiology and growth potential of poplars under short-rotation forestry culture. For. Ecol. Manag. 121, 9–23. Coleman, M.D., Isebrands, J.G., Tolsted, D.N., Tolbert, V.R., 2004. Comparing soil carbon of short rotation Poplar plantations with agricultural crops and woodlots in North Central United States. Environ. Manag. 33, 299–308. Copenheaver, C.A., Crawford, C.J., Fearer, T.M., 2011. Age-specific responses to climate identified in the growth of Quercus alba. Trees 25, 647–653. D’Amato, A.W., Bradford, J.B., Fraver, S., Palik, B.J., 2013. Effects of thinning on drought vulnerability and climate response in north temperate forest ecosystems. Ecol. Appl. 23, 1735–1742. Davis, S.C., Dietze, M., DeLucia, E., Field, C., Hamburg, S.P., Loarie, S., Long, S.P., 2012. Harvesting carbon from eastern US forests: opportunities and impacts of an expanding bioenergy industry. Forests 3, 370–397. DeBell, D.S., Harrington, C.A., 1997. Productivity of Populus in monoclonal and polyclonal blocks at three spacings. Can. J. For. Res. 27, 978–985. Don, A., Osborne, B., Hastings, A., Skika, U., Carter, M.S., Drewer, J., Zenone, T., 2011. Land-use change to bioenergy production in Europe: implications for the greenhouse gas balance and soil carbon. Glob. Chang. Biol. Bioenergy 4, 372–391. Fares, S., Mugnozza, G.S., Corona, P., Palahi, M., 2015. Five steps for managing Europe’s forests. Nature 519, 407–409. Farquhar, G., Caemmerer, S., Berry, J., 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C 3 species. Planta 149, 78–90. Fu, Y.H., Campioli, M., Van Oijen, M., Deckmyn, G., Janssens, I.A., 2012. Bayesian comparison of six different temperature-based budburst models for four temperate tree species. Ecol. Model. 230, 92–100. Grace, J.B., 2006. Structural Equation Modeling and Natural Systems. Cambridge
University Press, Cambridge, Massachusetts, USA. Grace, J.B., Bollen, K.A., 2008. Representing general theoretical concepts in structural equation models: the role of composite variables. Environ. Ecol. Stat. 15, 191–213. Hansen, E.A., 1993. Soil carbon sequestration beneath hybrid poplar plantations in the north central United States. Biomass Bioenerg. 5, 431–436. Heilman, P.E., Xie, F.G., 1993. Influence of nitrogen on growth and productivity of short-rotation Populus trichocarpa-Populus deltoides hybrids. Can. J. Res. 23, 1863–1869. Hill, J., Nelson, E., Tilman, D., Polasky, S., Tiffany, D., 2006. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. USA 103, 11206–11210. Hu, Y.L., Zheng, D.H., Fan, Z.P., Chen, G.S., Zhao, Q., Pepper, D., 2008. Changes in ecosystem carbon stocks following grassland afforestation of semiarid sandy soil in the southeastern Keerqin Sandy Lands, China. J. Arid Environ. 72 (12), 2193–2200. Kauter, D., Lewandowski, I., Claupein, W., 2003. Quantity and quality of harvestable biomass from Populus short rotation coppice for solid fuel use – a review of the physiological basis and management influences. Biomass Bioenergy 24, 411–427. Kline, R.B., 2005. Principles and Practice of Structural Equation Modeling, Second edition. Guilford Press, New York, USA. Kolström, M., Lindner, M., Vilén, T., Maroschek, M., Seidl, R., Lexer, M.J., et al., 2011. Reviewing the science and implementation of climate change adaptation measures in European forestry. Forests 2, 961–982. http://dx.doi.org/10.3390/ f2040961. Leonelli, G., Denneler, B., Bergeron, Y., 2008. Climate sensitivity of trembling aspen radial growth along a productivity gradient in northeastern British Columbia, Canada. Can. J. For. Res. 38, 1211–1222. Leuning, R., 1995. A critical appraisal of a combined stomatal photosynthesis model for C3 plants. Plant Cell Environ. 18 (4), 339–355. Liesebach, M., Wuehlisch, G.V, Muhs, H.-J., 1999. Aspen for short-rotation coppice plantations on agricultural sites in Germany: Effects of spacing and rotation length on growth and biomass production of aspen progenies. For. Ecol. Manag. 121, 25–39. Linares, J.C., Camarero, J.J., Carreira, J.A., 2010. Competition modulates the adaptation capacity of forests to climatic stress: insights from recent growth decline and death in relic stands of the Mediterranean fir Abies pinsapo. J. Ecol. 98, 592–603. Makeschin, F., 1994. Effects of energy forestry on soils. Biomass Bioenergy 6, 63–69. McLane, S.C., Daniels, L.D., Aitken, S.N., 2011. Climate impacts on lodgepole pine (Pinus contorta) radial growth in a provenance experiment. For. Ecol. Manag. 262, 115–123. Mead, D.G., 2005. Opportunities for improving plantation productivity. How much? How quickly? How realistic?. Biomass Bioenergy 28, 249–266. Meiresonne, L., Schrijver, A.D., Vos, B.D., 2006. Nutrient cycling in a poplar plantation (Populus trichocarpa Populus deltoides, Beaupre´) on former arable land in northern Belgium. Can. J. Res. 37, 141–155. Mérian, P., Lebourgeois, F., 2011. Size-mediated climate-growth relationships in temperate forests: a multi-species analysis. For. Ecol. Manag. 261, 1382–1391. Mitchell, C.P., Stevens, E.A., Watters, M.P., 1999. Short rotation forestry operations, productivity and costs based on experience gained in the UK. Ecol. Manag. 121, 123–136. Moorcroft, P.R., Hurtt, G.C., Pacala, S.W., 2001. A method for scaling vegetation dynamics: the ecosystem demography model (ED). Ecol. Monogr. 74, 557–586. Orwig, D.A., Abrams, M.D., 1997. Variation in radial growth responses to drought among species, site, and canopy strata. Trees 11, 474–484. Pan, C., Tajchman, S.J., Kochenderfer, J.N., 1997. Dendroclimatological analysis of major forest species of the central Appalachians. For. Ecol. Manag. 98, 77–87. Parton, W., Scurlock, J., Ojima, D., Scholes, R., Schimel, D., Gilmanov, T., Kirchner, T., Menaut, J.C., Seastedt, T., Moya, E.G., Kamnalrut, A., Kinyamario, J., 1993. Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide. Glob. Biogeochem. Cycles 7, 785–809. Pellegrino, E., Di Bene, C., Tozzini, C., Bonari, E., 2011. Impact on soil quality of a 10year-old short-rotation coppice poplar stand compared with intensive agricultural and uncultivated systems in a Mediterranean area. Agric. Ecosyst. Environ. 140, 245–254. Richter, D.D., Markewitz, D., Trumbore, S.E., Wells, C.G., 1999. Rapid accumulation and turnover of soil carbon in a re-establishing forest. Nature 400, 56–58. Saldarriaga, J.G., West, D.C., Tharp, M.L., Uhl, C., 1988. Long-term chronosequence of forest succession in the up-per Rio Negro of Colombia and Venezuela. J. Ecol. 76, 938–958. Sartori, F., Lal, R., 2006. Potential soil carbon sequestration and CO2 offset by dedicated energy crops in the USA. Crit. Rev. Plant Sci. 25, 441–472. Sartori, F., Markewitz, D., Borders, B., 2007. Soil carbon storage and nitrogen and phosphorus availability in loblolly pine plantations over 4 to 16 years of herbicide and fertilizer treatments. Biochemistry 84, 13–30. Sellers, P.J., 1992. Biophysical models of land processes. In: Trenberth, K.E. (Ed.), Climate System Modeling. Cambridge University Press, Cambridge, UK, pp. 451–490. Somerville, C., Young, H., Taylor, C., Davis, S.C., Long, S.P., 2010. Feedstocks for lignocellulosic biofuels. Science 329, 790. Stanturf, J.A., Oosten, C., Netzer, D.A., Coleman, M.D., Portwood, C.J., 2001. Ecology and silviculture of poplar plantations. In: Dickmann, D.I., Isebrands, J.G., Eckenwalder, J.E., Richardson, J. (Eds.), Poplar Culture in North America. NRC Research Press, Ottawa, pp. 153–206.
D. Wang et al. / Environmental Research 144 (2016) 88–95
Trumbore, S., 2006. Carbon respired by terrestrial ecosystems-recent progress and challenges. Glob. Chang. Biol. 12, 141–153. Tullus, A., Tullus, H., Vares, A., Kanal, A., 2007. Early growth of hybrid aspen (Populus wettsteinii Hä met-Ahti) plantations on former agricultural lands in Estonia. For. Ecol. Manag. 245, 118–129. Ulzen-Appiah, F., Briggs, R.D., Abrahamson, L.P., Bickelhaupt, D.H., 2000. Soil carbon pools in short-rotation willow (Salix dasyclados) plantation four years after establishment. In: Proceedings of Bioenergy 2000, Buffalo, NY, October 15–19. Van de Walle., I., Van Camp, N., Van de Casteele, L., Verheyen, K., Lemeur, R., 2007. Short-rotation forestry of birch, maple, poplar and willow in Flanders (Belgium) I. Biomass production after 4 years of tree growth. Biomass Bioenergy 31, 267–275. Viovy, N., Ciais, P., 2009. A Combined Dataset for Eecosystem Modelling. 〈http:// dods.extra.cea.fr/10 data/p529viov/cruncep/readme.htm〉. Walko, R.L., Band, L.E., Baron, J., et al., 2000. Coupled atmosphere–biophysics–hydrology models for environmental modeling. J. Appl. Meteorol. 39, 931–944.
95
Wang, D., LeBauer, D.S., Dietze, M.C., 2010. A quantitative review comparing the yield of switchgrass in monocultures and mixtures in relation to climate and management factors. Glob. Chang. Biol. Bioenergy 2, 16–25. Wang, D., LeBauer, D., Kling, G., Voigt, T., Dietze, M.C., 2013. Ecophysiological screening of tree species for biomass production: trade-off between production and water use. Ecosphere 4 (11), 138. Wang, D., LeBauer, D.S., Dietze, M.C., 2012. Predicting yields of short-rotation hybrid poplar (Populus spp.) for the United States through model–data synthesis. Ecol. Appl. 23, 944–958. Wang, D., Jaiswal, D., Lebauer, D.S., Wertin, T.M., Bollero, G.A., Leakey, A.D., Long, S. P., 2015. A physiological and biophysical model of coppice willow (Salix spp.) production yields for the contiguous USA in current and future climate scenarios. Plant Cell Environ. 38, 1850–1865. http://dx.doi.org/10.1111/pce.12556. Weih, M., 2004. Intensive short rotation forestry in boreal climates: present and future perspective. Can. J. For. Res. 34, 1369–1378.