AOBPreview originally published online on December 12, 2002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Annals of Botany 91: 319-327, 2003
© 2003 Annals of Botany Company
Simulation of Leaf Area Development Based on Dry Matter Partitioning and Specific Leaf Area for Cut Chrysanthemum
1 Wageningen University, Department of Plant Sciences, Horticultural Production Chains Group, Marijkeweg 22, 6709 PG Wageningen, The Netherlands
* For correspondence. Fax +31 (0) 317 484709, e-mail ep.heuvelink{at}wur.nl
Received: 31 July 2002; Returned for revision: 16 September 2002 Published electronically: 12 December 2002; Accepted: 18 October 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
This work aims to predict time courses of leaf area index (LAI) based on dry matter partitioning into the leaves and on specific leaf area of newly formed leaf biomass (SLAn) for year-round cut chrysanthemum crops. In five glasshouse experiments, each consisting of several plant densities and planted throughout the year, periodic destructive measurements were conducted to develop empirical models for partitioning and for SLAn. Dry matter partitioning into leaves, calculated as incremental leaf dry mass divided by incremental shoot dry mass between two destructive harvests, could be described accurately (R2 = 0·93) by a Gompertz function of relative time, Rt. Rt is 0 at planting date, 1 at the start of short-days, and 2 at final harvest. SLAn, calculated as the slope of a linear regression between periodic measurements of leaf dry mass (LDM) and LAI, showed a significant linear increase with the inverse of the daily incident photosynthetically active radiation (incident PAR, MJ m–2 d–1), averaged over the whole growing period, the average glasshouse temperature and plant density (R2 = 0·74). The models were validated by two independent experiments and with data from three commercial growers, each with four planting dates. Measured shoot dry mass increase, initial LAI and LDM, plant density, daily temperature and incident PAR were input into the model. Dynamics of LDM and LAI were predicted accurately by the model, although in the last part of the cultivation LAI was often overestimated. The slope of the linear regression of simulated against measured LDM varied between 0·95 and 1·09. For LAI this slope varied between 1·01 and 1·12. The models presented in this study are important for the development of a photosynthesis-driven crop growth model for cut chrysanthemum crops.
Key words: Chrysanthemum, dry mass partitioning, leaf area index, model, plant density, simulation, specific leaf area.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Explanatory models have an open modular character, which enables integration of knowledge at the level of underlying processes and transfer to other crops, as only some of the modules rather than the complete model may need adapting (Heuvelink, 1996). For glasshouse fruit vegetables, explanatory photosynthesis-driven models have been developed and thoroughly validated for the dynamics of dry mass production and dry mass partitioning (De Koning, 1994; Marcelis, 1994; Heuvelink, 1996). Contrary to vegetable crops, the number of photosynthesis-driven models for ornamental crops is very limited (Marcelis et al., 1998).
The accuracy of photosynthesis-driven models depends greatly on an accurate prediction of leaf area since crop growth is largely determined by intercepted light (Heuvelink, 1999). However, prediction of leaf area index (LAI) is still a weakness in these models. Two approaches are predominantly used to simulate plant leaf area development: (1) leaf area is described as a function of plant developmental stage; or (2) leaf area is predicted from simulated leaf dry mass (Marcelis et al., 1998). The former approach is often inaccurate for glasshouse crops owing to a large fluctuation in radiation (almost year-round cultivation), whereas leaf area development is often strongly influenced by radiation (Marcelis et al., 1998). Simulation of leaf area based on simulated leaf dry mass and specific leaf area (SLA) is a more flexible approach and one that has been applied to several crops, e.g. lettuce (Van Henten, 1994), tomato (Heuvelink 1999) and rose (Lieth and Pasian, 1991). A module for partitioning to the leaves and a module predicting SLA are needed to predict LAI dynamics in this way. In this approach, SLA is assumed to be constant or else is simulated as a function of plant age, physiological age, season, developmental stage and plant density or environmental conditions (Marcelis et al., 1998). Gary et al. (1995) calculated leaf area mainly as a function of temperature and physiological age. These authors distinguished between storage and structural leaf dry mass, and allowed structural SLA to vary between a minimum (full satisfaction of growth demand) and a maximum (minimum leaf thickness) value. This may be a promising approach for the mechanistic simulation of SLA and thus leaf area expansion; however, in this approach many parameters have to be estimated, and Gary et al. (1995) did not validate this part of their model.
In contrast to a tomato crop that showed constant partitioning to the leaves as a fraction of the total growth of vegetative parts (Heuvelink, 1999), this fraction was constant in chrysanthemum during the vegetative phase only (Acock et al., 1979). Partitioning to the leaves declined strongly as fraction of total plant growth and total vegetative growth during the generative phase of chrysanthemum (Hughes and Cockshull, 1971; Karlsson and Heins, 1992). For chrysanthemum, the SLA of new leaf biomass (SLAn, increase in LAI divided by increase in leaf mass) has been described by Acock et al. (1979) as a function of the average daily radiation integral and temperature. However, to the best of our knowledge, no attempt has been made to predict LAI for chrysanthemum during growth, although a promising approach might be to do this based on dry matter partitioning into the leaves and SLAn, based on Acock et al. (1979). Such a module for prediction of LAI would increase the applicability of photosynthesis-driven crop growth models for cut chrysanthemum (Heuvelink et al., 2001). Unfortunately, there is not only a lack of quantitative data on dry matter partitioning into the leaves and SLAn throughout the growing period of cut chrysanthemum crops, but also the empirical module for SLAn (Acock et al., 1979) is based on measurements taken in the vegetative phase and at high LAI only.
The aim of this study was to predict LAI development for cut chrysanthemum crops based on dry matter partitioning into the leaves and SLAn. An accurate prediction of dry matter partitioning into the leaves and SLAn is needed, as both have a strong positive feedback on LAI and total biomass production (e.g. Heuvelink, 1999). Five glasshouse experiments were conducted with different planting dates combined with several plant densities. These experiments were used to determine the general pattern of dry mass partitioning towards the leaves, to calibrate an existing empirical module for SLAn (Acock et al., 1979) and to develop a new module for predicting SLAn. Moreover, the model for LAI prediction was validated by two independent experiments and using data collected from commercial growers.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
General experimental set-up
Seven experiments were conducted in glasshouse compartments (12 x 12·8 m) of a multispan Venlo-type glasshouse at Wageningen University, The Netherlands (52°N), in different seasons in 1999 and 2000 (Table 1). Block-rooted cuttings of cut chrysanthemum [Chrysanthemum (Indicum group) ‘Reagan Improved’; CBA, Aalsmeer, The Nether lands] were obtained from a commercial propagator. Crops were planted on four or eight parallel soil beds (1·125 x 10·25 m per bed; a border soil bed was always present on both sides of the experimental soil beds) at three plant densities (32, 48 or 64 plants m–2) in two or three compartments. Five experiments were used to develop and calibrate models for dry matter partitioning into leaves and SLAn, and two experiments were used for model validation (Table 1).
|
General crop management was as described previously (Lee et al., 2002). Experiments were conducted in two (expts 1, 3 and 7) or three (expts 2, 4, 5 and 6) glasshouse compartments and, with the exception of expt 7, involved three plant densities (Table 1). Experiments 2–4 in this study are the same as expts 2, 6 and 4 described by Lee et al. (2002). Experiment 1 was conducted in parallel to expt 1 of Lee et al. (2002); however, two other glasshouse compartments were set at a higher glasshouse temperature of 21 °C. In expt 5 the crop was planted at 32, 64 and 80 plants m–2, but measurements taken at a density of 80 plants m–2 were ignored so as to keep plant densities the same for all experiments. In expt 6 each compartment half received supplementary lighting, provided by either incandescent lamps (4·8 ± 0·8 µmol m–2 s–1) or high-pressure sodium lamps (HPS, 39·6 ± 0·5 µmol m–2 s–1, SON-T 400W; Philips, Amsterdam, The Netherlands). In expt 7 three planting dates, spaced at weekly intervals, were used to expose plants to 3, 2 or 1 weeks of long-day (LD) conditions. The crop was planted at 64 plants m–2 and received supplementary assimilation light (HPS, 57·6 ± 0·8 µmol m–2 s–1).
LD conditions (16 h for 17–22 d after planting, expts 2 and 6; 19 h for 7–21 d, expts 1, 3 and 7) were provided by incandescent or HPS lamps. Plants in expts 4 and 5 received natural day lengths (about 15 h) for 21–22 d after planting (Table 1). Short days (SD) (10 h, expts 2 and 6; 11 h, expts 1, 3, 4, 5 and 7) were enforced by using a blackout screen until the end of an experiment. Lamps were on continuously during daylight hours of the LD and SD periods, except in expt 7 where the lamps were turned on when the global radiation outside the glasshouse was less than 150 W m–2 and turned off when it exceeded 250 W m–2.
Glasshouse climate
The glasshouse temperature set point for heating was 18 °C in the day and 19 °C at night, with the exception of expt 1 where day/night temperature set points were 20 /21 °C. The set-point temperature for ventilation was always 1 °C higher than that for heating. Measuring and recording of glasshouse climate data have been described previously (Lee et al., 2002). The CO2 concentration in each compartment was maintained between 350 and 400 µmol mol–1 by enriching with pure CO2. Average daily glasshouse temperature (averaged over 24 h) and CO2 concentration (averaged between 1000 and1600 h), and daily global radiation outside are presented in Table 1.
Daily photosynthetically active radiation inside the glasshouse compartment (incident PAR, MJ m–2 d–1) was calculated according to Lee et al. (2002), applying a glasshouse transmissivity for diffuse radiation of 0·49, measured on a cloudy day. Supplementary light (assumed to be 100 % diffuse) and light reduction by blackout screens in the SD period were taken into account in the calculation of daily incident PAR.
Plant measurements
In all experiments destructive measurements were carried out every 3 to 12 d until the final harvest. Samples were taken from five or six plants per experimental plot, excluding border plants in two rows on each side of a bed. Total leaf area (LI-3100 area meter; LI-COR, Lincoln, NB, USA) and fresh and dry (105 °C for 14 h in a ventilated oven) mass of leaves (including petioles), stems and flowers were measured. No measurements were conducted on roots.
Dry matter partitioning into leaves
The fraction of shoot dry mass partitioned to leaves was calculated as the increment of leaf dry mass divided by the increment of shoot dry mass between two adjacent destructive measurements; negative values were assumed to be zero (Kropff and Van Laar, 1993). The relationship between this fraction and relative time (Rt) (Karlsson and Heins, 1992) was described by a Gompertz curve:
F = C exp(–e[–B(Rt–M)])(1)
where F is the fraction of dry mass partitioned to the leaves, C represents the maximum fraction, B represents the steepness and direction of the curve and M represents Rt for the inflection point of the curve. For B > 0, the value of F will increase from zero to C with increasing Rt, whereas for B < 0, F decreases from C to zero. Rt was scaled from 0 to 2: Rt is 0 at the planting date, 1 at the start of SD, and 2 at final harvest. Rt values between 0–1 and 1–2 are obtained by linear interpolation of days after planting or start of SD. Parameters in eqn (1) were determined by the non-linear fitting procedure in the SPSS software package (version 10; SPSS company, Chicago, USA), applied to all partitioning data from expts 1–5.
Specific leaf area of new leaves
The specific leaf area of new leaves (SLAn, m2 g–1) is difficult to derive for each interval between two destructive measurements and hence the slope of the linear relationship between leaf dry mass (g m–2) and LAI (m2 m–2) was used to estimate SLAn (Kropff and Van Laar, 1993). This implies the assumption of a constant SLAn during each cultivation of chrysanthemum. An existing empirical model of SLAA [eqn (2)] of new leaves (Acock et al., 1979) was validated with estimated SLAn:
LA = a + bT + c/I(2)
where LA represents SLAA, I represents average daily incident PAR (400–700 nm; MJ m–2 d–1) and T average glasshouse temperature (°C), the latter two averaged over the whole growing period. Regression coefficients a, b and c have been estimated by Acock et al. (1979) as 5·23 m2 kg–1, 0·617 m2 kg–1 °C–1 and 43·74 m2 kg–1 MJ m–2 d–1, respectively. As conditions in the present experiments, including the cultivar, were quite different from those of Acock et al. (1979), regression parameters in eqn (2) were calibrated using the non-linear fitting procedure in SPSS. Furthermore, SLAn was predicted according to eqn (2), but extended with a linear term for plant density and with parameters calibrated on the SLAn values from expts 1–5.
Validation of the models
Periodic measurements of shoot dry mass (g m–2) from independent experiments (expts 6 and 7; Table 1) and from crops of commercial growers (Table 2) were fitted with a cubic spline function by Prism® (GraphPad Software Inc., San Diego, USA) to obtain daily crop growth rates (g m–2 d–1). Leaf growth rate (LGR, g m–2 d–1) was calculated by multiplying the partitioning to the leaves, calculated from eqn (1), with this daily crop growth rate. Leaf dry mass (LDM; g m–2) resulted from initial LDM and the cumulative LGR. The daily increase in LAI was calculated by multiplying predicted LGR (g m–2 d–1) by predicted SLAn (m2 g–1). LAI resulted from initial LAI and the cumulative daily increase in LAI. Initial values were input to the model and originated from destructive measurements at the time of planting.
|
Commercial crops for validation
Plants grown by three commercial growers were measured in four seasons in 2001, and measurements were conducted four times during a cultivation, i.e. at planting, at the start of SD, halfway through the SD period and at the commercial harvest stage (anthesis). Commercial crops were planted at densities between 40 and 62 plants m–2, and were grown at temperatures between 18 and 23 °C and CO2 concentrations between 380 and 1200 µmol mol–1 depending on the season (Table 2). The commercial glasshouses were Venlo-type glasshouses, but with a much higher transmissivity (68–70 %; measured on a cloudy day) and much bigger cultivation area (more than 1 ha) compared with the compartments in which the experiments were conducted. Cultivars ‘Reagan Elite White’ and ‘Reagan Elite Sunny’, used by the commercial growers, are very similar to ‘Reagan Improved’ used in the experiments; however, the period from the start of SD to final harvest is about 4 d shorter than for ‘Reagan Improved’ (CBA, Aalsmeer, The Netherlands).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Model development
A single Gompertz curve [eqn (1)] described well (R2 = 0·93) the relationship between dry mass partitioned to the leaves and relative time for all five experiments (Fig. 1A). In the early growth stages, dry mass partitioned to the leaves was 65 % of the total shoot dry mass, and this fraction was almost zero at anthesis. There was no statistically significant interaction between experiment and plant density, nor any effect of plant density alone (Fig. 1B), on any of the three parameters of the Gompertz curve. However, all three parameters were significantly affected by experiment (season). This effect can be seen in Fig. 1A: in general, measured fractions in expts 4 and 5 (open symbols) are below the curve, whereas fractions for expts 1–3 (closed symbols) are above the Gompertz curve.
|
A close linear relationship (R2 > 0·98) was observed between measured LDM and LAI during crop growth in each of the five experiments. The slope of these lines was taken as an estimate of SLAn. Applying eqn (2) with the parameter values determined by Acock et al. (1979) and the daily incident PAR and glasshouse temperature averaged over the whole growing period for each experiment resulted in a strong overestimation of SLAn for expts 1–3 (Fig. 2). For expts 4 and 5, predicted SLAn was almost equal to the values measured. Prediction was better if the regression parameters calibrated for the present experiments were used in the same equation (Fig. 2). The slope of the regression line (no intercept) relating predicted to measured SLAn was 0·995 (R2 = 0·51) indicating, on average, a perfect agreement but with much scatter. Following addition of a term for plant density (Pdm–2) to eqn (2), predictions changed little but scatter was much reduced [eqn (3); slope of regression line = 0·996; R2 = 0·67]. Therefore, SLAn (m2 kg–1) was calculated by:
|
Ln = a + bT + c/I + dPd(3)
where Ln represents SLAn, a = 3·99 ± 8·59 m2 kg–1, b = 0·989 ± 0·383 m2 kg–1 °C–1, c = 12·76 ± 1·5 m2 kg–1 MJm–2 d–1 and d = 0·0873 ± 0·0262 m2 kg–1 m2, accounting for 74 % of the variance.
Validation of leaf dry mass production and LAI
In expts 6 and 7, observed leaf dry mass (LDM, g m–2) production patterns over time could be predicted well (Fig. 3), based on the partitioning function from Fig. 1. Moreover, simulated dynamics of LAI in time also agreed well with measured patterns, although some overestimation at high density occurred in expt. 6 (Fig. 4). The slope of the linear relationship (no intercept) between measured and simulated LDM was 0·98 for expt 6 and 1·06 for expt 7, whereas for LAI these values were 1·12 for expt 6 and 1·01 for expt 7 (Table 3). Furthermore, by applying the simple relationship describing dry mass partitioning to the leaves as a function of relative time (Fig. 1) and the function describing SLAn as function of daily incident PAR, temperature and plant density [eqn (3)], observed and predicted LDM and LAI in 12 commercial crops (three growers x four planting dates) all showed very good agreement (Fig. 5; Table 4). Slopes of the linear relationship between predicted and observed LDM varied between 0·95 and 1·09, whereas for LAI these slopes varied between 1·01 and 1·11 (Table 4). Some overestimation of LDM and LAI was observed in summer and autumn, at the end of the cultivation (Fig. 5).
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Dry mass partitioning to the leaves
The fraction of dry mass partitioned to leaves has been reported to be constant during the early vegetative phase (Hughes and Cockshull, 1971; Acock et al., 1979), whereas this fraction decreases rapidly with floral development in chrysanthemum (Hughes and Cockshull, 1971). Our results also show such a pattern, which could be described accurately as a function of relative time by a Gompertz curve (Fig. 1). A similar approach describing dry mass partitioning to leaves was used by De Visser (1994) as a function of the developmental stage of onion, and by Tei et al. (1996) as a function of the day after emergence for onion and red beet.
Only a few studies report on dry mass partitioning into leaves in chrysanthemum. Acock et al. (1979) observed almost no influence of temperature (10–30 °C) or daily incident PAR (1·9–9·2 MJ m–2 d–1) on dry mass partitioning to leaves for a vegetative crop. However, Hughes and Cockshull (1971) and Karlsson and Heins (1992) reported that the fraction of dry mass partitioned to leaves increased with decreased light intensity, whereas the effect of CO2 concentration (Hughes and Cocksull, 1971) and day and night temperature (Karlsson and Heins, 1992) on this fraction seemed to be small.
In the present study, no significant effect of plant density on partitioning to the leaves was found (Fig. 1B). This finding is in agreement with observations of Reuben and Mnzava (1982) in Amaranthus cruentus. However, an effect of experiment (season) was observed on all three parameters of the Gompertz curve (Fig. 1A). This seasonal effect was not taken into account as it was preferred to keep the model simple. Furthermore, our glasshouse experiments do not allow a sound separation of the seasonal effect into a light and a temperature component, which would be needed for a generalization. Using a generalized pattern may lead to over- or underestimation of LDM and hence LAI. Overestimation of LAI was observed at the end of expts 4 and 5 (data not shown). Since LAI is high at the end of the cropping period, resulting in interception of almost all light (closed canopy), overestimation of LAI hardly influences light interception and hence simulated crop growth rate. Therefore, the general pattern for partitioning to the leaves was accepted without incorporating the specific effects of light, temperature or plant density, for example, which would have resulted in a more accurate prediction of LAI in the last weeks of cultivation. This general pattern of partitioning to leaves was validated by independent experiments and by data from several commercial growers. Even though there were large differences in crop management and environmental conditions between the present experiments and the commercially grown crops, predicted leaf growth based on measured shoot dry mass agreed very well with measured leaf growth (Fig. 5). Possible cultivar-specific effects have not been investigated here. However, the ratio between LDM and shoot dry mass has been reported to vary between 0·59 and 0·77 for 15 cultivars at a shoot dry mass of 1 g (De Jong and Jansen, 1992). Hence, the parameters of the partitioning curve (Fig. 1) are likely to be cultivar specific.
It may be that partitioning into the leaves can be described even more accurately by determination of a real developmental stage x-axis for Fig. 1. So far, only a rather rough linear interpolation of relative time between planting date, the start of SD and final harvest date was applied. The model for dry mass partitioning into the leaves is incomplete, as it needs final harvest date (anthesis) as input. The time from the start of SD until anthesis is primarily a cultivar characteristic. Larson and Persson (1999) attempted to predict the time to flowering from the start of SD. They noted that a lack of information from breeders is one of the main problems, since at least 25 new chrysanthemum cultivars become available every year. For known cultivars, one can also apply reference schedules, i.e. planting date, number of long days, expected final harvest date (e.g. Roelofs et al., 2001; Spaargaren, 2002) to obtain the time-axis for partitioning.
Specific leaf area
Specific leaf area of newly formed leaf dry mass (SLAn) has been predicted as a linear function of temperature and the inverse of daily light integral [eqn (2)] by Acock et al. (1979). However, eqn (2) with the parameter values of Acock et al. (1979) showed a large overestimation of SLAn for the present crops (expts 1–3; Fig. 2). This discrepancy can be explained by the fact that Acock et al. (1979) derived their equation for vegetative chrysanthemums only, under limited crop size (LAI > 2·2) and for relatively high incident PAR (1·9–9·2 MJ m–2 d–1). The latter may explain why predictions for the summer experiments (expts 4 and 5) were more accurate, as PAR levels were within the range of those of Acock et al. (1979). Since Acock et al. (1979) used a different cultivar to the one used in the present experiments, this may also explain differences in SLAn.
Using the calibrated parameters, eqn (2) showed good agreement with measured SLAn, and addition of a positive linear relationship with plant density [eqn (3)] improved prediction of SLAn further. The variation in SLA depends on light intensity or season, in agreement with published work (Hughes and Cockshull, 1972; Nederhoff et al., 1992; Heuvelink and Marcelis, 1996; Heuvelink, 1999). In addition, a small effect of temperature was found in chrysanthemum (Hughes and Cockshull, 1972). A positive effect of plant density on SLA has been found in other crops, e.g. potato (Vos, 1995), tomato (Heuvelink and Marcelis, 1996) and Impatiens capensis (Maliakal et al., 1999); this might be due to a reduction in average light levels on the leaves at higher plant densities. Furthermore, SLA decreases with increasing CO2 concentration (325–1500 µmol mol–1; Hughes and Cockshull, 1972), but CO2 concentration is not represented in eqn (3). However, using eqns (1) and (3) to simulate LAI, with total crop growth rate, initial LDM and initial LAI, temperature, incident PAR and plant density as input gave accurate predictions for crops grown in commercial glasshouses. The effects of [CO2] on SLAn cannot be large as commercial crops receive much higher CO2 concentrations (especially in winter) than did the plants used in the present experiments to develop the model. The parameter values in eqn (3) are likely to be cultivar specific: De Jong and Jansen (1992) observed that the ratio between leaf area and LDM varied between 0·0344 and 0·0468 m2 g–1 at a shoot dry mass of 1 g for 15 cut chrysanthemum cultivars grown in the same environmental conditions.
Many authors have attempted to predict LAI using a constant SLA, SLA as a function of developmental stage and day of the year, or sink/source relationships e.g. for tomato or rose (Marcelis et al., 1998). LAI is often hugely overestimated before canopy closure (Lieth and Pasian, 1991; Heuvelink 1999). In this study, prediction of LAI before canopy closure (LAI 
3) agreed very well with measurements (Figs 4 and 5B), using a simple approach based on dry mass partitioning to leaves as a function of relative time, and SLAn as a function of daily incident PAR, temperature and plant density. This approach is valuable for photosynthesis-driven crop growth models, as their accuracy depends strongly on prediction of light interception.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
Predicted LAI based on dry mass partitioning to leaves and SLA of new leaf biomass agreed well with measurements in validation experiments and with data from commercial growers. This approach, although demonstrated here for cut chrysanthemum crops, seems applicable to many other crops. Since no models were previously available for prediction of LAI dynamics in cut chrysanthemum crops, this work could be a valuable contribution towards a mechanistic photosynthesis-driven crop growth model for this crop. Heuvelink et al. (2001) introduced such a model, and the present results could be used to adjust and improve this existing model.
|
ACKNOWLEDGEMENTS |
|---|
We thank Westland Energie Services for offering commercial growers’ data and Professor Hugo Challa for valuable comments on this manuscript.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSLITERATURE CITED |
|---|
-
Acock B, Charles-Edwards DA, Sawyer S. 1979. Growth response of a Chrysanthemum crop to the environment. III. Effects of radiation and temperature on dry matter partitioning and photosynthesis. Annals of Botany 44: 289–300.
De Jong J, Jansen J. 1992. Genetic differences in relative growth rate and partitioning growth components in Chrysanthemum morifolium. Scientia Horticulturae 49: 267–275.[CrossRef]
De Koning ANM. 1994. Development and dry matter distribution in glasshouse tomato: a quantitative approach. PhD Thesis, Wageningen University, The Netherlands.
De Visser CLM. 1994. ALCEPAS, an onion growth model based on SUCROS87. I. Development of the model. Journal of Horticultural Science 69: 501–518.
Gary C, Barczi JF, Bertin N, Tchamitchian M. 1995. Simulation of individual organ growth and development on a tomato plant: a model and a user-friendly interface. Acta Horticulturae 399: 199–205.
Heuvelink E. 1996. Tomato growth and yield: quantitative analysis and synthesis. PhD Thesis, Wageningen University, The Netherlands.
Heuvelink E. 1999. Evaluation of a dynamic simulation model for tomato crop growth and development. Annals of Botany 83: 413–422.
Heuvelink E, Marcelis LFM. 1996. Influence of assimilate supply on leaf formation in sweet pepper and tomato. Journal of Horticultural Science 71: 405–414.
Heuvelink E, Lee JH, Carvalho SMP. 2001. Modelling visual product quality in cut chrysanthemum. Acta Horticulturae 566: 77–84.
Hughes AP, Cockshull KE. 1971. The effects of light intensity and carbon dioxide concentration on the growth of Chrysanthemum morifolium cv. Bright Golden Anne. Annals of Botany 35: 899–914.
Hughes AP, Cockshull KE. 1972. Further effects of light intensity, carbon dioxide concentration, and day temperature on the growth of Chrysanthemum morifolium cv. Bright Golden Anne in controlled environments. Annals of Botany 36: 533–550.
Karlsson MG, Heins RD. 1992. Chrysanthemum dry matter partitioning patterns along irradiance and temperature gradients. Canadian Journal of Plant Science 72: 307–316.
Kropff MJ, Van Laar HH. 1993. Modelling crop-weed interactions. Wallingford: CAB International.
Larsen RU, Persson L. 1999. Modelling flower development in greenhouse chrysanthemum cultivars in relation to temperature and response group. Scientia Horticulturae 80: 73–89.[CrossRef]
Lee JH, Heuvelink E, Challa H. 2002. Effects of planting date and plant density on crop growth in cut chrysanthemum. Journal of Horticultural Science and Biotechnology 77: 238–247.
Lieth JH, Pasian CC. 1991. A simulation model for the growth and development of flowering rose shoots. Scientia Horticulturae 46: 109–128.[CrossRef]
Maliakal SK, McDonnell K, Dudley SA, Schmitt J. 1999. Effects of red to far-red ratio and plant density on biomass allocation and gas exchange in Impatiens capensis. International Journal of Plant Sciences 160: 723–733.[CrossRef]
Marcelis LFM. 1994. Fruit growth and dry matter partitioning in cucumber. PhD Thesis, Wageningen University, The Netherlands.
Marcelis LFM, Heuvelink E, Goudriaan J. 1998. Modelling biomass production and yield of horticultural crops: a review. Scientia Horticulturae 74: 83–111.[CrossRef]
Nederhoff EM, De Koning ANM, Rijsdijk AA. 1992. Leaf deformation and fruit production of glasshouse grown tomato (Lycopersicon esculentum Mill.) as affected by CO2, plant density and pruning. Journal of Horticultural Science 67: 411–420.
Reuben SOWM, Mnzava NA. 1982. Dry matter partitioning in Amaranthus cruentus (L.) Thell. in response to population density. Beitrage zur tropischen Landwirtschaft und Veterinarmedizin 20: 131–134.
Roelofs T, Lee JH, Heuvelink E. 2001. Belichtingsniveau chrysant heeft optimum nog niet bereikt [Assimilation light intensity applied in practice in cut chrysanthemum is still sub-optimal]. Vakblad voor de Bloemisterij 18: 48–49.
Spaargaren JJ. 2002. De teelt van jaarrond chrysanten [Cultivation of year-round cut chrysanthemums]. Aalsmeer: Spaargaren.
Tei F, Scaife A, Aikman DP. 1996. Growth of lettuce, onion, and red beet. 1. Growth analysis, light interception, and radiation use efficiency. Annals of Botany 78: 633–643.
Van Henten EJ. 1994. Validation of a dynamic lettuce growth model for greenhouse climate control. Agricultural Systems 45: 55–72.
Vos J. 1995. The effects of nitrogen supply and stem density on leaf attributes and stem branching in potato (Solanum tuberosum L.). Potato Research 38: 271–279.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||