The Effects of Nitrogen Fertilization and the Growing Season on Carbon Partitioning in a Sward of Tall Fescue (Festuca arundinacea Schreb)

Annals of Botany 70 239244,1992 The Effects of Nitrogen Fertilization and the Growing Season on Carbon Partitioning in a Sward of Tall Fescue (Festuca arundinacea Schreb) G BELANGER*f, F GASTAL* and F R WAREMBOURG} * Station a"ecophysiologie des Plantes Fourrageres NRA, Centre de Recherches de Lusignan, 86600 Lusignan and % Centre L Emberger, CNRS, B P 5051, 34033, Montpelher Cedex, France Accepted 18 March 1992 The effect of N fertilization on the relative carbon partitioning to the roots of tall fescue {Festuca arundinacea Schreb), grown under field conditions, was studied with a 14 Clabellmg technique on three regrowths representing contrasting growing seasons Under nonlimiting N growing conditions, the relative carbon partitioning to the roots averaged 170, 15 8, and 11 1% in the summer, autumn, and spring regrowths, respectively The relative carbon partitioning to the roots increased during the summer and autumn regrowths but decreased during the spring regrowth n the absence of N fertilization, the relative carbon partitioning to the roots averaged 31 3,26 5, and 26 7 % in the summer, autumn, and spring regrowths, respectively The results were interpreted in terms of a functional equilibrium between the shoots and the roots t was concluded that, for a dense canopy of a perennial grass growing under fluctuating conditions of solar radiation and temperature, the relative growth of the roots compared to the relative growth of the total biomass is primarily a function of the shoot biomass Key words: Festuca arundinacea Schreb, carbon, partitioning, nitrogen, root growth, fertilization, grass NTRODUCTON Carbon partitioning between the roots and the shoots is one of the main determinants of shoot growth along with solar radiation interception and the conversion of intercepted solar radiation into plant assimilates (CharlesEdwards, 1982) Most studies conducted under controlled conditions or on fieldgrown spaced plants have shown that a N deficiency resulted in an increase in the relative carbon partitioning to the roots (Barta, 1975, Powell and Ryle, 1978, Robson and Parsons, 1978, Caloin, Khodre and Atry, 1980, Gastal and Saugier, 1986, Jarvis and Macduff, 1989) To our knowledge, nofieldstudies of the effect of N deficiency on carbon partitioning between the shoots and the roots in perennial grasses grown in a sward have been reported The growing season of perennial grasses is characterized by climatic variables such as temperature and incoming solar radiation (level and duration), and by the stage of plant development (vegetative or reproductive) Parsons and Robson (1981), working under high levels of N fertilization and at least 90% light interception, reported that the relative carbon partitioning to the roots of perennial ryegrass (Lolium perenne L ) decreased during autumn and remained relatively low during winter t increased early in the spring and then decreased during stem elongation Warembourg and Shakiba (1987) reported that the carbon allocation to the roots of Dactylis glomerata L grown under nonlimiting N and water supplies was maximum in early Present address Research Station, Agriculture Canada, Frederlcton, New Brunswick, Canada E3B 4Z7 spring and minimum during seed formation Warembourg and Paul (1977) found that the relative carbon partitioning to the roots of a permanent grassland reached its minimum in the spring when the crop was actively growing The effect of one or two climatic variables on carbon partitioning has been studied under controlled conditions The positive effect of low temperatures on the relative carbon partitioning to the roots has been reported on many species (Davidson, 1969, Pollock et al, 1983, Szaniawski, 1983, Shishido etal,\ 989) The relative carbon partitioning to the roots was reduced by low light intensity under controlled conditions (Ryle, 1970, Ryle and Powell, 1976, Gastal and Saugier, 1986) The study of carbon partitioning to roots of an established perennial grass cannot be achieved by measurements of the variation in root dry matter since this variation only represents the balance between growth, senescence and decomposition The carbon partitioning to the roots should therefore be studied with a tracer technique The objective of this work was to determine, under field conditions and using a 14 Clabelhng technique, the effect of N fertilization on the carbon partitioning coefficients to the roots The carbon partitioning coefficients were measured on three regrowths of tall fescue representing three contrasting growing seasons MATERALS AND METHODS The study was conducted at the Station d'ecophysiologie des Plantes Fourrageres de l'lnstitut National de la Recherche Agronomique, located in Lusignan, France (lat 03057364/92/090239 + 06 S08 00/0 1992 Annals of Botany Company

240 Belanger et al Nitrogen and Seasonal Effects on Festuca 46 26' N) Carbon partitioning to the roots of an established tall fescue (Festuca arundinacea Schreb cv Clanne) sward was studied on three contrasting regrowths summer (4 Jul16 Aug 1988), autumn (17 Oct6 Dec 1988) and spring (20 Mar3 May 1989) The experimental plots for each regrowth were treated uniformly prior to their use for the study and they received a relatively low rate of N fertilization A different field location was used for each regrowth The average daily global radiation was 19 8, 6 2, and 14 1 MJ m"! during the summer, autumn and spring regrowths, respectively The average daily temperature was respectively 18 1, 8 7, and 10 0 C The experimental plots were irrigated during the summer regrowth The water received by rainfall and irrigation during the summer regrowth was approximately equal to the potential evapotranspiration Three or four levels of N fertilization were used 0, 80, and 240 kg N ha" 1 in summer, 0, 40, and 160 kg N ha" 1 in autumn, 0, 60, 120, and 180 kg N ha" 1 in spring The N fertilization was done immediately after the defoliation preceding each regrowth period The highest levels of N fertilization in each regrowth were chosen so that the N content was not limiting shoot growth (Belanger, 1990) The carbon partitioning coefficients to the roots, also referred to as the relative carbon partitioning, were studied using a 14 Clabelling method adapted for sward conditions (Warembourg and Paul, 1973) The sward labelling was conducted with a closed chamber (50 x 50 x 70 cm) in which the CO 2 concentration and the temperature were maintained respectively at 330 + 25 /tmol mol" 1 and within ±1 C of the ambient temperature The sward was exposed to 14 COj for 1 d (photopenod) The specific activity of the solution used to generate the 14 CO 2 was 9 2, 18 5, and 4 6MBqg" 1 C during the summer, autumn, and spring regrowths, respectively For the summer and autumn regrowths, one level of N fertilization was applied on each side of an iron plate which was inserted vertically into the soil to a depth of 50 cm at least 3 weeks prior to the experimental period Swards grown with the two contrasting N levels were then labelled with 14 C on the same day, the labelling chamber covering both sides of the iron plate The possibility of confounding the effect of N fertilization with an effect of the day of 14 C labelling was therefore eliminated The 14 Clabelled area for each N level was 50 x 25 cm 14 Clabelling of the sward receiving no N fertilization was conducted on a different day using an area of 50 x 50 cm The results obtained in the summer and autumn regrowths indicated that the effect of the day of 14 Clabelhng on the relative carbon partitioning was negligible compared to the effect of N fertilization relabelling during the spring regrowth was therefore conducted on an area of 50 x 50 cm and on separate days for experimental plots receiving different rates of N fertilization The shoots, the shoot bases left below ground and the roots were harvested 7 d after labelling The shoots for each N level were harvested at ground level on an area of 15 x 40 cm Three soil cores of 8 cm in diameter were then taken to a depth of 60 cm The shoot bases left below ground were separated from the roots at their morphological connection after the roots had been washed free of soil The shoots, roots and shoot bases left below ground were dried at 80 C for 48 h, weighed and ground The 14 C was counted by liquid scintillation following dry combustion and CO 2 trapping (Bottner and Warembourg, 1976) The coefficients of carbon partitioning to a plant compartment were calculated as C t = CAJCA (0 where C k is the coefficient of carbon partitioning to a plant compartment k, CA t is 14 C activity in a plant compartment k, and CA is the M C activity in the total biomass (roots, shoots and shoot bases left belowground) The 14 C activity in a plant compartment and the total biomass were calculated by multiplying the specific activity of the plant compartment or the total biomass by their respective biomass Hence the coefficients of carbon partitioning to a plant compartment k can be expressed as SA t x B t SAxB where SA t is the specific activity of the plant compartment k (Bq mg C" 1 ), SA is specific activity of the total biomass (Bq mg C" 1 ), B t is biomass of plant compartment k (mg C), and B is total biomass (mg C) There were no replications of the experimental treatments since only one labelling chamber could be used The variability of the carbon partitioning coefficients for each N fertilization level and each labelling day was assessed by calculating the standard error among the three soil cores in the summer and autumn regrowths, and the six soil cores in the spring regrowth RESULTS The growing season integrates both the vegetation status (vegetative or reproductive) and the combination of climatic t 45 10 20 30 Days after defoliation (2) 40 50 FG 1 The effect of the growing season on the carbon paruuoning coefficients to the roots under nonlimiting N growing conditions (bars indicate one standard error) ( ) Summer, ( ) autumn, ( ) spring

to 40 35 30 25 20 15 10 5 0 S«45 "5 40 o _g 35 30 C 25 <P 8 20 t 10 o a. c 5 o _ o 0 45 40 35 30 25 20 15 10 5 A B _ c L 1 1 1 i z Belanger et al. Nitrogen and Seasonal Effects on Festuca 241 / 1 s * 3^^ * J X s 1 1 1 1 J 1 1 1 1 i i i 1 1 1. 10 20 30 40 50 Days after defoliation FG 2 The effect of the N fertilization rates on the carbon partitioning coefficients to the roots during the summer (A), autumn (B), and spring (C) rtgrowths (bars indicate one standard error) A, ( ) 240 kg N ha" 1, 160 kg N ha" 1, 180 kg N ha" 1, ( ) 80 kg N ha' 1, ( ) 0 kg N ha' 1 ) B, ( ) ( ) 40 kg N ha" 1, ( ) 0 kg N ha" 1 C, ( ) ( ) 120 kg N ha" 1, ( ) 60 kg Na ha" 1, ( ) conditions, the seasonal variations in soil N mineralization presumably resulted in swards having different N status even though similar amounts of N fertilization were applied Under nonlimiting N growing conditions, obtained with the highest rate of N fertilization for each regrowth, the coefficients of carbon partitioning to the roots ranged from 10 9 to 24 6 % with an average of 17 0 % during the summer regrowth (Fig 1) n the autumn regrowth, the coefficients of carbon partitioning to the roots ranged from 11 1 to 19 2% with an average of 15 8% The carbon partitioning coefficients ranged from 6 4 to 13 8% with an average of 11 1 % during the spnng regrowth The coefficients of carbon partitioning to the roots, averaged over the entire regrowth, were therefore highest in the summer and lowest in the spnng The relative carbon partitioning to the shoot bases left below ground averaged 6 5, 12 9, and 4 6 % during the summer, autumn, and spnng regrowths, respectively The change in the carbon partitioning coefficients to the roots dunng regrowth also differed from one growing season to another The proportion of assimilates exported to the root system increased dunng the summer and autumn regrowths The increase was, however, greater dunng the summer regrowth than dunng the autumn regrowth since the proportion of assimilates exported to the root system reached a plateau dunng the autumn regrowth The proportion of assimilates exported to the root system decreased dunng the spnng regrowth The N deficiency resulted in an increase in the relative carbon partitioning to the roots in all three regrowths (Fig 2) The carbon partitioning coefficients to the roots in the absence of N fertilization averaged 31 3, 26 5, and 26 7% in the summer, autumn, and spnng regrowths, respectively The relative carbon partitioning to the roots was 18, 17, and 2 4 times greater (summer, autumn, spnng) in the absence of N fertilization than under nonlimiting N growing conditions The carbon partitioning coefficients to the shoot bases left below ground in the absence of N fertilization were 7 3, 19 3, and 10 3% in the summer, autumn, and spnng regrowths, respectively DSCUSSON Results from studies conducted under controlled conditions indicated that "Clabelled assimilates required a maximum of approximately 5 d to reach their final destination as structural carbon and longterm storage (Michulnas et al, 1985, Danckwerts and Gordon, 1987, Belanger, 1990) Hence, since there was a penod of 7 d between u Clabelling and sampling, the relative carbon partitioning measured in this study refers to the carbon used for growth and longterm storage but it does not include the carbon lost by respiration vanables typical of each season The effect of the growing season could be analysed only in situations for which the N status of the sward in the three regrowths was equivalent This situation occurred only when N was not limiting shoot growth (Belanger, 1990) Under N deficient growing Nonlimitmg N growing conditions The range of carbon partitioning coefficients obtained under nonlimiting N growing conditions in the autumn and spnng regrowths was similar to those published by Parsons

242 Belanger et al Nitrogen and Seasonal Effects on Festuca and Robson (1981) on perennial ryegrass and by Warembourg and Shakiba (1987) on orchardgrass The carbon partitioning coefficients were also within the order of magnitude of the ratio of root biomass to total biomass measured on relatively young tall fescue plants grown under controlled conditions (Gastal and Saugier, 1986) The priority given to the shoots immediately after a defoliation has been interpreted in terms of a functional equilibrium between the roots and the shoots, and their respective activities (Brouwer, 1962) This functional equilibrium implies that the carbon assimilation rate is proportional to the N absorption rate, and that the plant reaches this equilibrium by adjusting the relative dimensions of the shoot and the root biomass (Wilson, 1988) During the summer and autumn regrowths, the shoots received a high proportion of carbon when the carbon assimilation of the sward was reduced following defoliation The decrease in the relative carbon partitioning to the roots during the spring regrowth was also reported for perennial ryegrass by Parsons and Robson (1981) and for orchardgrass by Warembourg and Shakiba (1987) n the present study, the average height of the growth apex was 10 cm 30 d after the start of the regrowth, indicating that stem elongation was under way The decrease in the relative carbon partitioning to the roots at the end of the spring regrowth period can therefore be ascribed to the increased shoot requirements for assimilates associated with stem elongation Many studies have indicated that low air temperature increased the proportion of assimilates allocated to the root system (Davidson, 1969, Pollock et al, 1983, Belanger, 1990) Our results, however, indicated that the relative carbon partitioning to the roots was less or equal in the autumn regrowth (average air temperature of 8 7 C) than in the summer regrowth (average air temperature of 18 1 C) The relative carbon partitioning to both the roots and the shoot bases left below ground averaged over the entire regrowth was greater in the autumn regrowth (28 7%) than in the summer regrowth (23 5 %) This result emphasizes the importance of the shoot bases as a compartment for carbon allocation during the autumn regrowth, presumably originating from fructan storage in existing tillers and the production of new tillers The relative tillering rate was higher during the autumn regrowth than during the spring and summer regrowths (data not shown) Ndefictent growing conditions The N deficiency caused an increase in the proportion of carbon exported to the root system and to the shoot bases left below ground in all three regrowths This result, obtained under field conditions and with a dense canopy, is in agreement with those obtained on different species grown under controlled conditions (Brouwer, 1962, Robson and Parsons, 1978, Powell and Ryle, 1978, Caloin et al, 1980, Gastal and Saugier, 1986) This result may also be viewed as a confirmation of the functional equilibrium A higher priority is given to the roots when the N absorption is reduced under Ndeficient growing conditions Relative specific activity of the shoots From eqn (2), the carbon partitioning coefficient to a plant compartment k can be seen as the product of the relative specific activity (RSA) of that plant compartment and the ratio between the biomass of plant compartment k and the total biomass SA B * B The relative specific activity is the ratio between the specific activity of a plant compartment and the specific activity of the total biomass t is a dimensionless number used to compare the 14 C concentration of a specific sink or plant compartment with that in the total biomass on a unit mass basis (Craddle and Heichel, 1988) The relative specific activity therefore represents the strength of a sink or a plant compartment relative to the total biomass n terms of dry matter change, the carbon partitioning coefficient can be expressed as (3) C t = ABJAB (4) By combining eqns (3) and (4), it is shown that the relative specific activity is equal to the ratio of the relative growth of a plant compartment and the relative growth of the total biomass RSA t = AB/B (5) When the data obtained over the three regrowths were pooled, it was found that the shoot relative specific activity (RSA a ) decreased with an increase in the shoot biomass (B a ) (Fig 3) Fitting allometnc equations to the data gave the following relationships RSA a = 114 8(5J* 78 r 1 = 0 70 for no N, (6) RSA a = 61 6(B a Y r 1 = 0 76 for nonlimiting N (7) Compared to the total biomass and per unit of carbon, the shoots used a decreasing proportion of the newly assimilated carbon when the existing amount of carbon in the shoot increased n terms of relative growth, those relationships indicate that the ratio of the shoot relative growth to the total biomass relative growth decreased as the shoot biomass increased The decrease of the shoot relative specific activity with an increase in shoot biomass is described by an allometnc model similar to the one used to describe the decrease in N content with an increase in shoot biomass (Lemaire and Salette, 1984, Greenwood et al, 1990) The average leaf age increases during regrowth and the average radiation level received by the leaves decreases Hence the physiological activity of the shoots relative to that of the total biomass decreases with an increase in shoot biomass during regrowth t was found that the decrease in shoot relative specific activity was slightly greater in the absence of N fertilization than when N was not limiting growth Hence, the ratio of the relative growth of the shoots to the relative growth of the total biomass decreased faster during regrowth when N

Belanger et al Nitrogen and Seasonal Effects on Festuca 243 100 200 300 Shoot biomass (g DM m 2 ) 400 500 FG 3 The relationship between the shoot relative specific activity and the shoot biomass for nonlimiting N growing conditions ( ) and no N fertilization (O) (curves represent the allometnc equations and their 95% confidence intervals) was deficient than when N was not limiting growth The relationship between the shoot relative specific activity and shoot biomass, established for nonlimiting N growing conditions, describes the optimal behaviour of grass canopy during a regrowth in terms of relative growth of the shoots in relation to the total biomass When N is deficient, the grass canopy behaves similarly immediately after defoliation, but it deviates from the optimal behaviour as shoot biomass accumulates A linear relationship between the relative growth rate of a compartment and the relative growth rate of the total biomass was predicted for isolated plants with exponential growth (CharlesEdwards, Doley and Rimmington, 1986) This result was obtained analytically from a model of carbon partitioning based on a functional equilibrium Our results indicate that the relative growth of the shoots in relation to the relative growth of the total biomass decreased with an increase in shoot biomass The ratio of the relative growth rates of different plant compartments was therefore not constant as was suggested by CharlesEdwards et al (1986) The predictions made by CharlesEdwards et al (1986) were based on the assumption that the specific shoot activity was not a function of the shoot weight This assumption is reasonable with young, wellspaced plants as suggested by CharlesEdwards et al (1986) but it does not hold true for a plant canopy Our results, however, indicate that the model of carbon partitioning based on a functional equilibrium could be applied to the data obtained on perennial grasses in natural conditions if the shoot biomass was taken into account t is concluded that, for a dense canopy of a perennial grass growing under fluctuating conditions of solar radiation and temperature, shoot biomass is the main determining factor of the shoot relative specific activity For any N fertilization level, the shoot relative specific activity or the shoot relative growth compared to the total biomass relative growth is always high when the shoot biomass is low as in the early part of the regrowth As shoot biomass accumulates, the shoot relative specific activity decreases for all N fertilization but it decreases faster when N is limiting growth LTERATURE CTED Barta AL. 1975. Effect of nitrogen nutrition on distribution of photosynthetically incorporated 14 CO, in Lolium perenne Canadian Journal of Botany 53 237 242 Bllanger G 1990. ncidence de la fertilisation azotee et de la saison sur la croissance, Vassimilation et la repartition du carbone dans un couvert de fetuque e'levee en conditions naturelles Thesis (D Sc ), Universite de PansSud, Pans Bottner P, Warembourg FR. 1976. Method for simultaneous measurement of total and radioactive carbon in soils, soils extracts and plant materials Plant and Soil 45 273277 Brouwer R 1962. Nutritive influences on the distribution of dry matter in the plants Netherlands Journal of Agricultural Science 10 399408 Caloin M, Kbodre AE, Atry M. 1980. Effect of nitrate concentration on the root/shoot ratio in Dactylis glomerata L and on the kinetics of growth in the vegetative phase Annals of Botany 46 165173 CharlesEdwards DA. 1982. Physiological determinants of crop growth Sydney Academic Press CharlesEdwards DA, Doley D, Rimmington GM. 1986. Modelling plant growth and development Sydney Academic Press Craddle HT, Heicbel GH. 1988. Photosynthate partitioning in alfalfa before harvest and during regrowth Crop Science 28 948953 Danckwerts JE, Gordon AJ 1987. Longterm partitioning, storage and remobilization of U C assimilated by Lolium perenne (cv Melle) Annals of Botany 59 5566 Davidson RL 1969. Effects of root/leaf temperature differentials on root/shoot ratios in some pasture grasses and clover Annals of Botany 33 561569 Gastal F, Saugier B. 1986. Alimentation azotee et croissance de la fetuque elevee 1Assimilation du carbone et repartition entre organes Agronomie 6 157 166 Greenwood DJ, Lemalre G, Gosse G, Cruz P, Draycott A, Neeteson JJ. 1990. Decline in percentage N of C3 and C4 crops with increasing plant mass Annals of Botany 66 425 436 Jarvis SC, Macdnff JH. 1989 Nitrate nutrition of grasses from steadystate supplies in flowing solution culture following nitrate deprivation and/or defoliation Journal of Experimental Botany 40 965975 Lemaire G, Salette J 1984. Relation entre dynamique de croisance et dynamique de prelevement d'azote pour un peuplement de graminees fourrageres 1 Etude de l'effet du milieu Agronomie 4 423430 Michulnas DG, Laurenroth WK, Singh JS, Cole CV, Hunt HW. 1985. Root turnover and production by "Cdilution implications of carbon partitioning in plants Plant and Soil 88 353365 Parsons AJ, Robson MJ. 1981. Seasonal changes in the physiology of S24 perennial ryegrass (Lolium perenne L) Partition of assimilates between root and shoot during the transition from the vegetative to reproductive growth Annals of Botany 48 733744 Pollock CJ, Uoyd EJ, Stoddart JL, Thomas H 1983. Growth, photosynthesis and assimilate partitioning in Lolium temulentum exposed to chilling temperatures Plant Physiology 59 257262 PoweU CE, Ryie GJA. 1978. Effect of nitrogen deficiency on photosynthesis and the partitioning of u Clabelled assimilates in unshaded and partially shaded plant of Lolium lemulentum Annals of Applied Biology 90 241248 Robson MJ, Parsons AJ. 1978. Nitrogen deficiency in small closed communities of S24 ryegrass Photosynthesis, respiration, dry matter production and partition Annals of Botany 42 11851197 Ryle GJA. 1970. Partition of assimilates in an annual and a perennial grass Journal of Applied Ecology 1 217227 Ryte GJA, Powell CE. 1976. Effect of rate of photosynthesis on the pattern of assimilate distribution in the graminaceous plant Journal of Experimental Botany 27 189199 ShishWo V, Seyama N, mada S, Horl Y. 1989. Carbon budget in tomato plants as affected by night temperature evaluated by steady state feeding with "CO, Annals of Botany 63 357367

244 Belanger et al Nitrogen and Seasonal Effects on Festuca Szaniawski RK. 1983. Adaptation and functional balance between in grassland plant production and turnover, soil and plant shoot and root activity of sunflower plants grown at different root respiration Soil Biology and Biochemistry 9 295301 temperatures Annals of Botany 51 453459 Warembourg FR, Shakiba MR. 1987 Distribution of assimilated Warembourg FR, Paul EA. 1973. The use of 14 CO, canopy techniques carbon and production in a Dactylis glomerata L field ntecol for measuring carbon transfer through the plantsoil system Plant Bulletin 15 7582 and Soil 38 331345 Wilson JB. 1988. A review of evidence on the control of shoot root Warembourg FR, Paul EA 1977. Seasonal transfers of assimilated "C ratio, in relation to models Annals of Botany 61 433449