Nitrogen resorption in senescing leaf blades of rice exposed to free-air CO2 enrichment (FACE) under different N fertilization levels
Nitrogen (N) resorption from senescing leaves is essential to meet N demand for grain development in rice (Oryza sativa L.). We asked whether rice is capable of reducing N in their senesced leaf blade to lower concentration at elevated [CO2] more in low than in high N fertilization.
The effects of elevated [CO2] and N fertilization on senesced leaf N concentration were examined for 3 years with the free-air CO2 enrichment (FACE) technology.
Elevated [CO2] decreased the senesced leaf N concentration but the change was generally small and did not hold over the growing seasons. Additionally, there was no evidence that the change was greater at low than at high N fertilization levels.
The 3-year field measurements showed that elevated [CO2] did not change the senesced leaf N concentration consistently. The occasional decrease in senesced leaf N concentration was associated with a decrease in green leaf N concentration at elevated [CO2] but not with the proportion of leaf N resorbed during leaf senescence.
KeywordsAtmospheric carbon dioxide concentration Global change Leaf senescence Litter production Litter quality Nitrogen retranslocation
Increase in nitrogen (N) supply is expected to enhance rice production at elevated atmospheric carbon dioxide concentration ([CO2]). For example, a free-air CO2 enrichment (FACE) experiment conducted in northern Japan demonstrated that elevated [CO2] stimulated the yield to a greater extent at high- than low N fertilization (Kim et al. 2001, 2003). However, the economic and environmental costs of N fertilizer are high because its production uses considerable energy from fossil fuels and additionally, an excess N fertilization on soil is becoming a concern because of its environmental effects through leaching into groundwater and denitrification (e.g., Yang et al. 2004; Bouwman et al. 2013; Zhang et al. 2015). Thus, the input of N fertilizer into the soil may not be expected to increase in future.
Nitrogen required for grain development is supplied from soil and also from senescing tissues, in particular from leaves (Mae and Ohira 1981; Oritani 1984; Oritani and Yoshida 1984). It was suggested that N resorbed from vegetative organs accounts for 70–90% of N in the panicle and that N resorption is essential for the yield (Mae 1986, 1997). A large N resorption can contribute to grain development by supplying more N and also by improving N use efficiency for growth. The ability of N resorption from senescing leaves has been assessed using N resorption proficiency that is measured as the level to which leaf N concentration is reduced during senescence: i.e., lower litter N concentration is identified to be more proficient at resorbing N (Killingbeck 1996). Many studies have investigated the effects of elevated [CO2] on it, most of them reporting that the senesced leaf N concentration decreased at elevated [CO2] and the decreases were related to the decreases in N concentration of green leaf and not to the proportion of leaf N resorbed during leaf senescence (Norby et al. 2000; Billings et al. 2003; Norby and Iversen 2006; Housman et al. 2012).
To our knowledge, no attempts have been done to study the influence of elevated [CO2] on senesced leaf N concentration of rice in the paddy field. Meanwhile, decrease in green leaf N concentration at elevated [CO2] has been widely observed (Kim et al. 2003; Borjigidai et al. 2006; Yang et al. 2006; Chen et al. 2014). In addition, a greenhouse experiment found that the green leaf N concentration of rice decreased at elevated [CO2] more at low- than high N fertilization (Nakano et al. 1997). If the proportion of leaf N resorbed during leaf senescence does not respond to elevated [CO2] as in other species, it is hypothesized that senesced leaf N concentration of rice would decrease at elevated [CO2] more in low than in high N fertilization.
In this study, we investigate the influence of elevated [CO2] on senesced leaf N concentration in paddy rice at different N fertilization levels. The following questions are addressed. First, does senesced leaf N concentration decrease at elevated [CO2] more at lower- than at higher N fertilization? The measurements were taken over three growing seasons (2010–2012) as senesced leaf N concentration has been suggested to vary intra- and inter-annually in a single species (Killingbeck 1996; Drenovsky et al. 2013). Second, when senesced leaf N concentration is analyzed as the function of green leaf N concentration and the proportion of leaf N resorbed during leaf senescence, which of the two is more important to changes in senesced leaf N concentration at elevated [CO2]? This question is answered combining our results and the data on the green leaf N concentration obtained at the same experimental site in 2012 by Chen et al. (2014). Nitrogen that is not resorbed returns to soils later on as litter fall. We also determine leaf litter production and estimate the N in leaf litter both at ambient and at elevated [CO2].
Materials and methods
This study was conducted at the FACE facility in Tsukubamirai, Ibaraki, central Japan (35o58’N, 139 o59’W, 10 m above sea level). The FACE facility consisted of four blocks, each containing two 17 m diameter octagonal plots. Within each block, one plot was at ambient [CO2] and the other was fumigated with an elevated [CO2] of 200 μmol mol−1 above ambient. CO2 fumigation continued during daytime. In the experimental period, the actual daytime [CO2] averaged over a growing season with day-to-day standard deviation in the ambient [CO2] control plot was 385 ± 11.0 μmol mol−1 in 2010, 379 ± 13.9 μmol mol−1 in 2011, and 383 ± 11.2 μmol mol−1 in 2012, whereas that in the elevated [CO2] plot was 585 ± 15.9 μmol mol−1 in 2010, 560 ± 26.3 μmol mol−1 in 2011 and 578 ± 15.7 μmol mol−1 in 2012 (see Nakamura et al. 2012 for a full description of the FACE system). The mean air temperature during the experiment was 24.9 °C, 24.1 °C, and 23.7 °C, in 2010, 2011, and 2012, respectively. The total rainfall was 188 mm, 455.7 mm, and 443.7 mm, respectively.
Details of the cultural practices are given elsewhere (Hasegawa et al. 2013; Zhang et al. 2013; Usui et al. 2016). Briefly the treatment areas received an equal amount of P and K: 4.36 g m−2 of P and 8.30 g m-2 of K were applied as a PK compound fertilizer before flooding (early April). Each [CO2] plot was divided into subplots. Three levels of N were supplied to each subplot: 0 g N m−2 (LN), 8 g N m−2 (MN, a regional standard) and 12 g N m−2 (HN). The size of each subplot was larger than 8.1 m2. At MN and HN, 25% of N was supplied as urea, 50% as two types of control release fertilizer (types LP100 and LP140; JCAM Agri. Co., Tokyo, Japan) just before puddling, about a week prior to transplanting. Each N subplot was enclosed with PVC corrugate boarding placed vertically into the hardpan layer of the soil to prevent from lateral movement of N between subplots. The soil was a Fluvisol, typical of alluvial areas, whose total C and N contents were 21.4 mg C g−1 and 1.97 mg N g−1, respectively (see Hasegawa et al. 2013 for details). Nitrogen mineralization of the air-dried soils incubated at 30 °C for 30 days was approximately 160 μg N g−1 (T. Tokida, unpublished data). A commercial cultivar of rice (O. sativa L. cv. Koshihikari) was used. Seeds were sown in seedling trays on April 26 in 2010, April 25 in 2011, April 24 in 2012. Seedlings grown in a nursery under ambient [CO2] conditions were hand-transplanted in groups of three plants (referred to as a hill) into the experimental plots at a density of 22.2 hills m−2 (15 cm × 30 cm) on May 26–27 in 2010, May 25–26 in 2011 and May 23–24 in 2012.
We collected fully senesced leaf blades that were produced in 2 hills per subplot at 2-week-intervals from the early tillering stage through to the grain maturity stage. We determined the length of all senesced leaf blades and estimated the surface area using allometric relations between the length and area (r2 = 0.73–0.98) at each sampling; the relationships varied significantly among sampling dates but not among treatments (data not shown). After drying for 48 h in an oven set at 70 °C, the dry mass was determined and subsequently the samples were ground for N analyses. Nitrogen concentration was analyzed with an NC analyzer (Sumigraph NC-22F, Sumika-Bunseki, Osaka, Japan). N concentration of fully senesced leaf blades was determined both on a blade mass basis and on a blade area basis. Senesced leaf N concentration was related to green leaf N concentration and the proportion of leaf N resorbed (rN) as:
The values of green leaf N concentration (on a blade area basis) were obtained from Chen et al. (2014), who determined them at the MN subplot at around the heading stage (64, 72 and 77 days after transplanting [DAT]) in 2012. Senesced leaf N concentration (on a blade area basis) at these dates was estimated by interpolating between two adjacent observations of the senesced leaf N concentration. For the calculation of rN, we assumed that green leaf N concentration does not depend on the time of leaf emergence.
Litter production (dry mass of fully senesced leaf blade per unit ground area) was determined by harvesting 6 to 10 hills per subplot at MN and 4 hills per subplot at LN and at HN at the panicle initiation stage (41–45, 39–40 and 41–42 DAT in 2010, 2011 and 2012, respectively), the heading stage (68–70, 73–74 and 74–75 DAT in 2010, 2011 and 2012, respectively), and the grain maturity stage (101–105, 103–108 and 107–109 DAT in 2010, 2011 and 2012, respectively). Nitrogen in leaf litter was estimated as the product of the litter production and its respective senesced leaf N concentration that was obtained on adjacent days.
The effects of [CO2] elevation and N fertilization on N concentration of leaf blades, leaf litter production and N in leaf litter were analyzed with a randomized complete block split-plot mixed model analysis of variance. In all tests, [CO2] and N fertilization were fixed effects and plot was random effect. The effect of [CO2] elevation on the proportion of leaf N resorbed was analyzed with [CO2] as fixed effect and block as random effect. The calculations were performed with R (R Development Core Team, 2006). Relationship between N concentration in green- and fully senesced leaf blades was analyzed with standard major axis (SMA, Warton et al. 2006) using SMATR package for R.
Nitrogen resorption proficiency
Split-plot ANOVA of the effects of elevated CO2 treatment, nitrogen fertilization treatment and two-way interactions on nitrogen per unit area (NL,A) and nitrogen per unit mass (NL,M) of fully senesced leaf blade of rice in three growing seasons (2010–2012). Significant differences between treatments within each date (DAT) are denoted by asterisk; ***, ** and * indicate a significance level of P < 0.001, P < 0.010 and P < 0.050, respectively
F statistics, P value
9.22, 0.002 **
13.42, <0.001 ***
21.26, <0.001 ***
12.23, <0.001 ***
CO2 x N fertilization
4.71, 0.04 *
6.82, 0.007 **
CO2 x N fertilization
5.89, 0.03 *
15.34, <0.001 ***
23.99, <0.001 ***
55.85, <0.001 ***
58.48, <0.001 ***
41.90, <0.001 ***
39.71, <0.001 ***
CO2 x N fertilization
4.98, 0.02 *
5.96, 0.03 *
11.32, <0.001 ***
21.86, <0.001 ***
23.72, <0.001 ***
32.59, <0.001 ***
36.83, <0.001 ***
CO2 x N fertilization
7.00, 0.007 **
15.62, <0.001 ***
12.58, <0.001 ***
31.84, <0.001 ***
40.60, <0.001 ***
58.33, <0.001 ***
CO2 x N fertilization
5.62, 0.02 *
11.23, <0.001 ***
8.72, <0.001 ***
18.37, <0.001 ***
18.65, <0.001 ***
CO2 x N fertilization
Nitrogen concentration of fully senesced leaf blade expressed on an area basis (NL,A) was high at the start of experiment, decreased towards the panicle initiation stage and thereafter it was more or less constant (Fig. 1b) as NL,M decreased but the mass per unit area of the senesced leaf blade increased (Electronic Supplementary Material). Like NL,M, the inter-annual variation in NL,A was small. Influences of elevated [CO2] and N fertilization on NL,A were also similar to those on NL,M. Significant decrease in NL,A at elevated [CO2] was observed at 73–74 DAT in 2012. A decrease at elevated [CO2] was also found at HN and LN at 55 DAT in 2011 but not at MN, indicated by the significant interaction between [CO2] and N fertilization (Table 1). Nitrogen fertilization tended to increase NL,A though the difference between MN and LN was not always significant in 2010 and 2012.
Relationship between senesced and green leaf N
When the NL,A decreased at elevated [CO2], the green leaf N concentration (NG,A) decreased too (Chen et al. 2014) and the proportion of leaf N resorbed during leaf senescence (rN) did not differ between [CO2] levels (0.91 and 0.90 at ambient and elevated [CO2], respectively). The proportion of leaf N resorbed that was calculated from blade mass-based N concentration was also similar at both [CO2] levels (0.77 and 0.74 at ambient and elevated [CO2], respectively). Thus, when the senesced leaf N concentration decreased at elevated [CO2], it was due to the decrease in the green leaf N concentration.
Nitrogen in leaf litter
Leaf litter production and leaf litter nitrogen that were accumulated from the transplanting to the grain maturity stage, and results of the split-plot ANOVA to test for the effects of growth [CO2], N fertilization and their interaction
Values ± SD
(F statistics, P value)
CO2 x N
Leaf litter production (g DM m-2)
45.8 ± 16.6
63.3 ± 13.9
55.6 ± 11.8
39.6 ± 10.1
63.3 ± 14.4
57.9 ± 12.3
5.44, 0.02 *
69.7 ± 16.2
102.6 ± 22.8
86.4 ± 16.9
71.0 ± 10.2
121.7 ± 17.2
101.4 ± 17.5
5.94, 0.03 *
13.10, <0.001 ***
56.1 ± 9.1
94.8 ± 14.0
107.2 ± 32.8
52.4 ± 12.5
94.4 ± 6.5
118.2 ± 11.9
31.47, <0.001 ***
Leaf litter nitrogen (g N m-2)
0.23 ± 0.10
0.32 ± 0.10
0.32 ± 0.09
0.19 ± 0.05
0.34 ± 0.04
0.33 ± 0.07
6.16, 0.01 *
0.24 ± 0.06
0.41 ± 0.06
0.44 ± 0.07
0.24 ± 0.02
0.48 ± 0.08
0.46 ± 0.10
23.45, <0.001 ***
0.20 ± 0.03
0.38 ± 0.05
0.52 ± 0.15
0.19 ± 0.04
0.37 ± 0.02
0.54 ± 0.05
70.11, <0.001 ***
While we determined the senesced leaf N concentration on both mass- and area basis, many other studies used N concentration per unit leaf mass and examined how environmental or genetic factors influenced N resorption (e.g., Kobe et al. 2005; Yuan and Chen 2009; Hayes et al. 2014). They assumed that resorption of carbon and nutrients other than N was not influenced by those factors. Area-based N of leaf litter does not require such assumption but is often hard to obtain because leaf area decreases (i.e., shrinks) during leaf senescence. We avoided this problem by estimating the area of senesced blade from the blade length. N concentration on area basis showed a similar pattern to that on mass basis.
The senesced leaf N concentration did not change with [CO2] level consistently over the growing seasons. This does not agree with Strain and Bazzaz (1983) who first hypothesized that N concentration of green tissues and litter decreases at elevated [CO2], and also with earlier field experimental studies (Norby et al. 2000; Billings et al. 2003; Norby and Iversen 2006; Housman et al. 2012). Various factors other than [CO2] could confound the effect of elevated [CO2]. For example, defoliation by herbivores and early frost may inhibit N resorption from damaged leaves and thereby increase variation in senesced leaf N concentration within a treatment (Killingbeck 1996). Genetic difference and soil N heterogeneity can also increase the variation among individuals of a single species (Lü et al. 2011). In the rice paddy, their effects would be minimal because herbivory was severely controlled with pesticide applications and the sampling was finished by the late summer before frost occurrence. In addition, the flooded water and the soil puddling homogenized the soil N distribution, and genetic difference of the modern rice cultivar could be small (Ram et al. 2007). Nonetheless, the [CO2] effect on senesced leaf N concentration was occasionally detected by the repeated measurements. Therefore, it would be concluded that the influence of [CO2] on the ability of rice to reduce the N in senescing leaf blade is inherently small, if any.
In 2010, the senesced leaf N concentration did not change with [CO2] levels throughout the growing season, while in 2011 and 2012, it decreased at elevated [CO2] around the heading stage (Table 1, Fig. 1). The difference in response among years might be associated with difference in environmental factors. For example, high air temperature and low rainfall were characteristic in 2010 (see Materials and Methods). In Ambrosia dumosa growing in the Mojave desert, the senesced leaf N concentration decreased at elevated [CO2] in an average- or a high-rainfall year but did not change in a low-rainfall year (Housman et al. 2012). In Lycium pallidum, on the other hand, the senesced leaf N concentration was not influenced by elevated [CO2] in all years. Hence, the difference in senesced leaf N response among years does not appear to be simply explained by changes in temperature and rainfall, although studies are lacking in mesic ecosystems.
Around the heading stage, the senesced leaf N concentration decreased at elevated [CO2], which was associated with a decrease in the green leaf N concentration (Fig. 2) but not with the proportion of leaf N resorbed (rN). Decrease in green leaf N concentration at elevated [CO2] around the heading stage was also observed in previous FACE experiments of rice (Kim et al. 2003; Borjigidai et al. 2006). In rice, the decrease in green leaf N concentration at elevated [CO2] is explained by a decrease in N allocation to leaf blades but not by leaf N dilution (Makino and Mae 1997). No study has so far examined CO2-response of rN in paddy rice but our observation was consistent with those in many other ecosystems, where rN did not change with [CO2] elevation (Norby et al. 2000; Finzi et al. 2001; Billings et al. 2003; Housman et al. 2012). Thus it seems to be a general phenomenon.
The rN was higher when it was calculated from area- rather than mass-based leaf blade N. This is because carbon and nutrients other than N are also resorbed from senescing leaves and thus leaf mass decreases during leaf senescence (van Heerwaarden et al. 2003a). Irrespective of the basis, the rN of rice was much higher than previous data compilations (0.50, Aerts and Chapin 2000; 0.62, Vergutz et al. 2012). The high ability of N resorption may be a distinctive characteristics of rice. Because rN and leaf lifespan determine residence time of N (Aerts and Chapin 2000; Hirose and Oikawa 2012), and because the leaf lifespan of rice is relatively short due to its annual habit (Ogawa et al. 2016), the high rN would be essential for N use in rice. On the other hand, the rN could be unduly overestimated on some grounds. First, part of leaf blade N can be emitted to atmosphere as ammonia (NH3) if NH3 concentration was higher in leaf than in air. However, Hayashi et al. (2011) found that the rate of NH3 emission from the flag leaf was less than 12.9 g N ha−1 h−1 during the reproductive phase, and that there was no significant influence of [CO2] elevation on the NH3 emission rate. NH3 was not only emitted from leaf but also absorbed by leaf, and the balance between emission and absorption changed depending on the growth stages (Miyazawa et al. 2014). Second, we used the N concentration of green and senesced leaf blade to calculate rN, both of which were taken at around the heading stage. But the N concentration that the senesced leaf blades had achieved when they were green would be lower than the N concentration of green leaf blade present at the heading stage, as described above. When a 3.7% lower green leaf N concentration is assumed for the leaves that were shed at the heading stage compared to the leaves that matured at the heading stage, measured in a different cultivar of rice (O. sativa L. cv. Eiko; Tanaka 1956), rN becomes 0.91 at ambient [CO2] and 0.90 at elevated [CO2] in the current experiment; and thus the overestimation of rN would be very small at both treatments (<0.4%).
Senesced leaf N concentration increased with increasing N fertilization in all years (Fig. 1), suggesting that the costs of decreasing N concentration in the senesced tissues might be relatively higher than the cost of absorbing N from the soil (Wright and Westoby 2003). In contrast, decreasing senesced leaf N concentration should be important at N-deficient soils to conserve N and increase its residence time (e.g., Vitousek 1982; Killingbeck 1996). In fact, in many natural and semi-natural ecosystems, plants in infertile soils retained less N in the litter, compared to those in fertile soils (van Heerwaarden et al. 2003b; Richardson et al. 2005; Norris and Reich 2009; Inagaki et al. 2011; Hayes et al. 2014). In the present study, however, the difference in senesced leaf N concentration was not always significant between MN and LN (standard and no N fertilization, respectively). The lack of response might suggest that the senesced leaf N concentration reached the biochemical limit to which N can be reduced. In Japan, N fertilization to paddy soils is moderately controlled to avoid herbivory and pathogen infection, lodging and deteriorations of canopy structure (Matsushima 1995) and to increase the commercial value of grains (palatability) (Okadome et al. 1999; Wada et al. 2006). Thus the paddy rice might be exposed to N deficiency even when they were standardly fertilized. The fact that the values of senesced leaf N concentration at MN were close to the low end of senesced leaf N concentration variation in a data compilation for >1000 observations from many natural and crop species (Yuan and Chen 2009) may support this speculation.
The 3-year field measurements showed that elevated [CO2] did not change the senesced leaf N concentration consistently. This is different from the results in other ecosystems, where elevated [CO2] decreased senesced leaf N concentration. Our results suggest that a further increase in N input from outside is required to enhance rice production at elevated [CO2]. When the N concentration in leaf litter decreased, it was always associated with that in the N concentration in green leaf blade, without change in the proportion of leaf N resorbed during leaf senescence.
We gratefully acknowledge the staff members of NIAES for their support in operating and maintaining the Tsukubamirai FACE experimental facility. Kazuki Nakamura, Takahiro Ogawa and the other members of TUA provided assistance with sampling at the field. We also thank Kentaro Hayashi, Shin-Ichi Miyazawa and Toshihiko Kinugasa for comments. This work was supported in part by the Ministry of Agriculture, Forestry and Fisheries, Japan, through a research project entitled “Development of Technologies for Mitigation and Adaptation to Climate Change in Agriculture, Forestry and Fisheries”, in part by a Grant-in-Aid for Scientific Research on Innovative Areas (no. 21114009, 24114711) by the Japan Society for the Promotion of Science, as part of the project entitled, “Comprehensive Studies of Plant Responses to a High CO2 World by an Innovative Consortium of Ecologists and Molecular Biologists.”
SO and THi designed the research. HN and THa operated and managed the FACE experimental facility. SO, HE, MK, CPC, THa, HS, YU and TT conducted fieldwork. Laboratory work was carried out by SO, HE and MK. SO analyzed data and drafted the manuscript. THi, KH, TT, CPC, YU and THa contributed to manuscript revisions.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- Aerts R, Chapin FS III (2000) The mineral nutrition of wild plants revisited: a re-evaluation of processes and patterns. Adv Ecol Res 30:1–67Google Scholar
- Bouwman L, Klein Gldewijk K, Van Der Hoek KW, Beusen AHW, Van Vuuren DP, Wilems J, Rufino MC, Stehfest E (2013) Exploring global changes in nitrogen and phosphorus cycles in agriculture induced by livestock production over the 1900-2050 period. Proc Natl Acad Sci U S A 110:20882–20887CrossRefPubMedGoogle Scholar
- Finzi AC, Allen AS, Delucia EH, Ellsworth DS, Schlesinger WH (2001) Forest litter production, chemistry, and decomposition following two years of free-air CO2 enrichment. Ecology 82:470–484Google Scholar
- Hasegawa T, Sakai H, Tokida T, Nakamura H, Zhu C, Usui Y, Yoshimoto M, Fukuoka M, Wakatsuki H, Katayanagi N, Matsunami T, Kaneta Y, Sato T, Takakai F, Sameshima R, Okada M, Mae T, Makino A (2013) Rice cultivar responses to elevated CO2 at two free-air CO2 enrichment (FACE) sites in Japan. Func Plant Biol 40:148–159CrossRefGoogle Scholar
- Inagaki M, Kamo K, Titin J, Jamalung L, Lapongan J, Mura S (2011) Nutrient dynamics through fine litterfall in three plantations in Sabah, Malaysia, in relation to nutrient supply to surface soil. Nut Cycli Agr 88:381–395Google Scholar
- Kobe RK, Lepczyk CA, Iyer M (2005) Resorption efficiency decreases with increasing green leaf nutrients in a global data set. Ecology 86:2780–2792Google Scholar
- Mae T (1986) Partitioning and utilization of nitrogen in rice plants. Japan Agr Res Quarterly 20:115–120Google Scholar
- Mae T, Ohira K (1981) The remobilization of nitrogen related to leaf growth and senescence in rice plants (Oryza sativa L.). Plant Cell Physiol 22:1067–1074Google Scholar
- Matsushima S (1995) Physiology of high-yielding rice plants from the viewpoint of yield components. In: Matsuo T, Kumazawa K, Ishii R, Ishihara K, Hirata H (eds) Science of the rice plants, volume II. Food and Agriculture Policy Research Center, TokyoGoogle Scholar
- Miyazawa S-I, Hayashi K, Nakamura H, Hasegawa T, Miyao M (2014) Elevated CO2 decreases the photorespiratory NH3 production but does not decrease the NH3 compensation point in rice leaves. Plant Cell Physiol 55:1482–1591Google Scholar
- Norby RJ, Iversen CM (2006) Nitrogen uptake, distribution, turnover, and efficiency of use in a CO2-enriched sweetgum forest. Ecology 87:5–14Google Scholar
- Oritani T (1984) Studies on nitrogen metabolism in crop plants. XX. Translocation and accumulation into sink of 15N top-dressed at different growth stages in the rice plant. Jpn J Crop Sci 53:276–281 (In Japanese with English summary)Google Scholar
- R Development Core Team (2006) R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. ISBN 3–900051–07-0, URL http://www.R-project.org.
- Strain BR, Bazzaz FA (1983) Terrestrial plant communities. In: Lemon ER (ed) CO2 and plants. Westview Press, Boulder, pp 177–222Google Scholar
- Tanaka A (1956) Studies on characteristics of physiological function of leaf at definite position on stem of rice plant. Part 3. Relation between nitrogen metabolism and physiological function of leaf at definite position. Jpn J Soil Sci Plant Nut 26:27–32 (In Japanese with English summary)Google Scholar
- Zhang G, Sakai H, Tokida T, Usui Y, Zhu C, Nakamura H, Yoshimoto M, Fukuoka M, Kobayashi K, Hasegawa T (2013) The effects of free-air CO2 enrichment (FACE) on carbon and nitrogen accumulation in grains of rice (Oryza sativa L.). J Exp Bot 64:3179–3188Google Scholar