Sowing date affected shoot and root biomass accumulation of lucerne during establishment and subsequent regrowth season

Europ. J. Agronomy 68 (2015) 69–77

Contents lists available at ScienceDirect

European Journal of Agronomy journal homepage: www.elsevier.com/locate/eja

Sowing date affected shoot and root biomass accumulation of lucerne during establishment and subsequent regrowth season Richard E. Sim a , Derrick J. Moot a,∗ , Hamish E. Brown b , Edmar I. Teixeira b a b

Faculty of Agriculture and Life Science, PO Box 7647, Lincoln University, Canterbury, New Zealand The New Zealand Institute for Plant & Food Research Limited, Private Bag 4704, Christchurch, New Zealand

a r t i c l e

i n f o

Article history: Received 1 October 2014 Received in revised form 17 April 2015 Accepted 22 April 2015 Available online 15 May 2015 Keywords: Alfalfa Photoperiod Reserves Seedling Visible bud

a b s t r a c t The pattern of perennial dry matter (DM) was manipulated over two seasons to determine if the establishment of lucerne (Medicago sativa L.) is regulated by the demand for assimilate by perennial organs, (taproot plus crown) or crop ontogeny. Crops of ‘Stamina 5’ lucerne were established from spring to late summer at two sites which differed by 230 mm to 2.3 m soil depth in plant available water content (PAWC) at Lincoln University, New Zealand. The establishment phase was characterised from sowing until crops reached a maximum accumulation of perennial biomass of ∼5 t DM ha−1 . Demand for biomass offered insight into the variability in fractional partitioning of DM to the perennial organs (Proot ) during establishment. This showed that Proot was 0.48 until a perennial biomass of 2.9 ± 0.28 t DM ha−1 . Lucerne continued to partition DM to the perennial organs until a maximum biomass of ∼5 t DM ha−1 , but at a decreasing rate shown by a linear decline in Proot in response to increasing perennial biomass. This meant Proot was independent of crop ontogeny, but most likely still under the control of environmental influences, and the establishment phase extended into the second season for crops which had not attained a perennial biomass >3 t DM ha−1 . These crops continued to prioritise the allocation of DM to the perennial organs which explained the 20–25% decrease in shoot yield in the second season when sowing was delayed. This study quantified the establishment phase of lucerne to perennial biomass demand as independent of crop ontogeny. It showed establishment was regulated by biomass demand of these perennial organs. The spring sown crops on the High PAWC soils completed this phase at the earliest in 4 months. In contrast, autumn sown crops on the Low PAWC soils took nearly 9 months to complete this phase. These results indicate different management strategies may be required to establish lucerne rather than solely using first flowering as a sign that the establishment phase is complete. Results can be incorporated into the current partitioning framework to improve the simulation modelling of lucerne. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Crop ontogeny of lucerne (Medicago sativa L.) can be described by two distinct phases (i) seedling and (ii) regrowth. Agronomically, the seedling phase refers to the first growth cycle from sowing to first defoliation, which is recommended to occur at flowering (Moot et al., 2003). In contrast, regrowth phases are the growth cycles that occur following the initial and subsequent defoliation events. Alternatively, lucerne growth could be considered as an establishment phase (where perennial biomass is accumulated preferentially) followed by an established phase (where partitioning of biomass responds to environmental factors). It is important to understand the establishment phase to enable optimal stand management for

∗ Corresponding author. Tel.: +64 3 423 0705; fax: +64 3 325 3850. E-mail addresses: [email protected] (R.E. Sim), [email protected] (D.J. Moot). http://dx.doi.org/10.1016/j.eja.2015.04.005 1161-0301/© 2015 Elsevier B.V. All rights reserved.

maximum productivity and persistence (Fick et al., 1988). Furthermore, it may increase the accuracy of yield prediction of simulation models, which currently define lucerne phases in relation to crop ontogeny (Robertson et al., 2002). The current recommendation is to delay first defoliation until open flower buds are observed (Moot et al., 2003). This allows maximum opportunity to establish reserves of carbon and nitrogen in the perennial organs which consist of the taproot plus crown (Justes et al., 2002; Khaiti and Lemaire, 1992; Thiebeau et al., 2011). These reserves are used to support lucerne regrowth and survival during winter (Teixeira et al., 2007a; Volenec et al., 1996). During the seedling phase, biomass is preferentially partitioned to the perennial organs as the development of the root system is a stronger sink for assimilates (Khaiti and Lemaire, 1992). As a consequence of reduced assimilate available to the shoot, these crops experience up to half the leaf area expansion rate of regrowth crops (Teixeira et al., 2011). Therefore, less intercepted radiation is available for growth,

70

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

and seedling crops accumulate less shoot dry matter (Teixeira et al., 2011; Thiebeau et al., 2011). In contrast, the allocation of DM in regrowth crops is dynamic and the fractional partitioning of DM to the perennial organs (Proot ) ranges from ∼0.10 to 0.50 (Khaiti and Lemaire, 1992; Teixeira et al., 2008; Thiebeau et al., 2011) and seasonal variation has been described in relation to photoperiod and temperature (Teixeira et al., 2008). In most temperate regions, lucerne can be sown from spring to early autumn. Delayed sowing reduces intercepted radiation by the crop in the establishment season and shoot yields decrease accordingly (Teixeira et al., 2011; Thiebeau et al., 2011). Time of sowing studies, where shoot yield has been reported over two years, shows that shoot yields in the second season can also be influenced by the time of sowing in the previous year (Justes et al., 2002; Moot et al., 2012; Teixeira et al., 2011; Thiebeau et al., 2011). Moot et al. (2012) reported the shoot yields of autumn sown crops were ∼20% less in the first season after establishment, compared with their year 2–5 mean yield. Thiebeau et al. (2011) reported a similar yield reduction with delayed sowing from early spring to late summer, and showed this was due to preferential partitioning of DM to the perennial organs in the following spring, until a taproot biomass of 3–4 t DM ha−1 was reached. Our hypothesis is that the phase where DM is preferentially partitioned to the perennial organs is not limited only to the seedling phase. Instead crop establishment may extend further in response to the sink demand of the perennial organs for assimilates. The aims of this study were to (i) determine the length of lucerne establishment when DM allocation to the perennial organs is prioritised, and (ii) to identify if this phase is regulated by biomass demand or crop ontogeny. To do this, lucerne was sown in the field on 10 dates from spring to late summer over two years. Shoot yield and taproot plus crown biomass were quantified for the seedling and subsequent regrowth crops. To uncouple the influence of seedling phenology from DM assimilation and partitioning, growth rates were further manipulated by sowing into soils with contrasting amounts of plant available water.

2. Materials and methods 2.1. Experimental design, treatments and establishment Dryland lucerne was established as a split-plot within a randomised complete block design, replicated four times. Sowing date was the main plot and rhizobia inoculation carrier the sub plots (4.2 × 7 m). The experiment was replicated at two sites (Lincoln University main campus and Ashley Dene Research Farm) at the same latitude, but with two distinct soil types (high and low plant available water capacity). In the first growing season (2010/2011), lucerne was sown at monthly intervals on five dates from October to February, and this was repeated in the second year (Table 1). There were four carriers of rhizobia, Sinorhizobium meliloti (i) lime coated, (ii) peat, (iii) ALOSCA and (iv) bare seed control. The results of that experiment have been reported elsewhere (Khumalo et al., 2012; Wigley et al., 2012). Lime coated seed sub plots were sown in both years and are the focus of this study. Lime coated seed contains rhizobia (S. meliloti) and fungicide, which protect against Pythium spp. with additional molybdenum and lime. Prior to the first sowing date, both experimental areas were conventionally cultivated (plough, maxi-til, harrow and roll) and ‘Stamina 5’ lucerne seed was sown at a rate of 10.5 kg ha−1 (bare seed equivalent) with 92% germination using an Øyjord cone seeder. All crops established >200 plants m−2 within one month after sowing. Sequential sowing date treatments created crops which differed in perennial biomass at the end of the

Table 1 Sowing dates for lucerne sown into soils of high and low plant available water capacity (PAWC) at Lincoln University, Canterbury, New Zealand. Season

Sowing date number

High PAWC

Low PAWC

2010/2011

1 2 3 4 5

4 October 4 November 2 December 10 January 7 February

21 October 9 November 8 December 13 January 3 February

2011/2012

6 7 8 9 10

10 October 7 November 9 December 10 January 17 February

10 October 7 November 9 December 10 January 17 February

first year, while the second season sowing enabled a comparison of seedling and regrowth lucerne in the same year. 2.2. Site characteristics and meteorological conditions The first site was at the Lincoln University main campus, located in paddock 12 of Iversen fields (43◦ 38 S, 172◦ 28 E, and is 11 m.a.s.l). The soil is classified as a Wakanui silt loam (Udic Ustochrept, USDA Soil Taxonomy), (Cox, 1978), which generally has 2–3 m of fine textured silt loam overlying gravels (Webb, 2003). The Ashley Dene Research Farm site was located in paddock M2B (43◦ 38 S, 172◦ 19 E, and is 30 m.a.s.l). The soil is a Lismore stony silt loam (Udic Haplustept loamy skeletal, USDA Soil Taxonomy), (Cox, 1978), which has a shallow topsoil (∼0.2 m) that contains ∼10% stones by volume, overlaying coarse gravels with a stone content up to 45% (Di and Cameron, 2002; Webb, 2003). Meteorological data were recorded at Broadfields Meteorological Station (NIWA, National Institute of Water and Atmosphere Research, New Zealand), 2 km north of the High PAWC site. Rainfall and air temperature were recorded at the experimental sites. Mean monthly temperature was consistent between sites and ranged from 6 ◦ C in July to 17 ◦ C in January. Mean monthly total solar radiation ranged from 4 to 22 MJ m−2 day−1 over the same period. Long term mean (50 year) annual rainfall is 630 mm, distributed evenly throughout the year, but ranged from 580 mm at the Low PAWC site to 645 mm at the High PAWC site in 2011/2012. In 2010/2011, rainfall was 610 mm at both sites. Long term mean annual Penman potential evapotranspiration (EP) is 1095 mm. Monthly total EP ranges from 35 mm in June/July to 150 mm in December/January and usually exceeds rainfall from September to April. Annual EP was 975 mm and 910 mm in 2010/2011 and 2011/2012, respectively. Photoperiod was calculated by time of year and site latitude and included civil twilight (Goodspeed, 1975). Photoperiod at this latitude ranges from 10.0 h on 21 June to 16.7 h on 22 December. 2.3. Agronomic management 2.3.1. Crop management Soil fertility was maintained to optimum levels (Morton and Roberts, 1999) through annual chemical analysis of the topsoil (0–150 mm), each autumn and applications of superphosphate (9% P, 11% S) and potassic sulphur super (8% P, 5% K, 10% S) as required (Sim, 2014). Plots were kept weed free by chemical control and hand weeding. 2.3.2. Defoliation Seedling lucerne, defined as the growth phase from sowing to first defoliation was cut when 50% of 10 marked stems per plot had an open flower (Section 2.4.3). During the autumn months, seedling lucerne was defoliated when 50% of the marked stems had a visible flower bud as the onset of frosts meant further reproductive devel-

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

opment was unlikely. Subsequent defoliations of regrowth lucerne occurred when 50% of marked stems had a visible flower bud. The exceptions were the first spring defoliation of lucerne in its second year. This occurred when crop height reached 0.35–0.40 m to prevent lodging, and was before reproductive development was initiated for crops at the High PAWC site. Summer regrowth cycles at the Low PAWC site were defoliated before visible flower bud stage due to the onset of severe moisture stress which resulted in rapid senescence of lower leaves. For one regrowth cycle during late summer, defoliation was delayed to allow 50% of marked stems to have an open flower. A final defoliation occurred once growth stopped in May/June. This management resulted in crops in the first year being defoliated 1–4 times depending on sowing date and 5–7 times in the second year at the Low and High PAWC sites, respectively. Defoliations were undertaken mechanically using a lawn mower to a height of 50 mm. The exception was at the Low PAWC site, where second year crops were grazed with approximately 100 ± 20 ewes for a duration of 7–10 days. 2.4. Measurements 2.4.1. Determining plant available water Volumetric soil water content (VWC) was measured in 22 layers of the soil profile to a depth of 2.3 m at 10 to 14 day intervals. The top layer (0–0.2 m) was measured with a time domain reflectometer (TDR; Trace system, Soil Moisture Equipment, Santa Barbara, California, USA) with 0.2 m long stainless steel rods. The remaining 21 layers were measured at their mid-point using a neutron probe (Troxler Electronic Industries Inc., Research Triangle Park, North Carolina, USA). Neutron probe access tubes were installed to a depth of 2.3 m in the centre of each coated seed sub plot. Due to the stony soil profile at the Low PAWC site, it was impossible for the conventional installation method to auger through the compacted gravels. Thus, spiked access holes were created using a 50 mm steel spike and a vibrating head attachment on a 20 tonne excavator. Aluminium tubes were then installed. To maintain consistency, this method was used at both experimental sites. Previous experience in a similar soil type using this installation method showed no effect on actual soil moisture (Mills, 2007). Access tubes were installed 5–7 days following each sowing date, to allow mechanical sowing of the entire plot, but prevent damage to seedling lucerne. PAWC was determined for each individual soil layer for each plot by the difference between the drained upper limit (DUL) and lower limit (LL) of water extraction. DUL was defined as the maximum stable VWC which was measured 5 days after complete soil recharge, to allow for drainage. Complete recharge was known to occur in the plots which were sown in the second season as they were chemically fallowed for the first season, incurring ∼600 mm of rainfall and no plant water extraction. The LL was identified as the lowest measured VWC for the mature crop in the second season which had explored soil moisture to a depth of at least 2.3 m. This showed the High PAWC site had 360 mm of water available to lucerne to a depth of 2.3 m (Fig. 1a) and the Low PAWC site had 130 mm of water available to the same depth (Fig. 1b). This contrast in PAWC allowed the comparison of shoot and root growth amongst crops grown in similar environments, but with different growth potential due to available soil water content. 2.4.2. Shoot and taproot plus crown biomass Shoot biomass was measured at 7–14 day intervals using a single 0.2 m quadrat, cut just above crown height (∼50 mm) using hand shears. Perennial biomass, consisting of the plant crown and taproot was excavated to 300 mm from a 0.2 m quadrat at the end of each season (June 2011 and 2012) and also at the end of each growth cycle for crops sown in the second season (sowing dates

71

6–10). This procedure is justified by the fact that 80–90% of total perennial biomass is sampled at this depth (Fornasier et al., 2003; Khaiti and Lemaire, 1992; Lemaire et al., 1992). Taproots were separated from the crowns and were washed free of soil. All DM samples were dried in a forced air oven (60 ◦ C) to a constant weight. 2.4.3. Reproductive development Reproductive development was recorded every 3–7 days from 10 marked stems per plot during the seedling phase and 5 marked stems during subsequent regrowth cycles. Stems were selected on different plants with an intentional bias to mark taller, dominant stems which have been shown to account for >80% of shoot yield (Teixeira et al., 2007b), and because shorter stems often senesce as the canopy develops. Reproductive development occurred when 50% of the marked stems were observed with visible buds or open flowers. In the 2011/2012, season across all crops, visible buds were displayed in 25 individual regrowth cycles compared with 8 which flowered. Therefore, only visible bud data are presented. Thermal time (Section 2.5.2) requirement to reach 50% visible buds was calculated from emergence for seedling crops and from defoliation for regrowth crops. 2.5. Data analysis 2.5.1. Growth rate Average daily growth rate for each growth cycle was calculated from the linear regression between regrowth period (days) and shoot DM accumulation. Growth rates were presented for sowing dates 1–5 because these included two years of data. In most cases, regressions were fitted to all data points, except when senescence occurred at the end of a growth cycle, and then these points were excluded. Growth rate data are presented for the mid point of each growth cycle. 2.5.2. Thermal time Thermal time (Tt, ◦ Cd) was calculated following the method of Jones and Kiniry (1986) from daily mean air temperature using a broken-stick threshold model, where Tt is assumed zero below the base temperature (Tb ) of 1.0 ◦ C. Tt is accumulated linearly at a rate of 0.7 ◦ Cd ◦ C−1 up to 15 ◦ C and then at a rate of 1.0 until 30 ◦ C (Moot et al., 2001; Teixeira et al., 2011). This method calculates Tt at three hourly intervals which are integrated over a day. 2.5.3. DM partitioning to taproot plus crown The fractional DM partitioned to taproot plus crown (Proot ) for an individual growth cycle was calculated from the final shoot yield and perennial biomass at the end of the growth cycle for sowing dates 6–10. The taproot plus crown biomass was calculated less the initial biomass at the start of the growth phase. At the end of the season, sometimes regrowth cycles were too short to enable recharge of perennial reserves and resulted in negative Proot values. This occurred only twice at each site and values were not included in the analysis. 2.5.4. Statistics Statistical analysis was conducted in GENSTAT (version 14.1) (Lawes Agricultural Trust, IACR, Rothamsted, UK). Analysis of variance (ANOVA) was used to determine differences amongst treatment means. Shoot yield and perennial biomass were compared for seedling and regrowth crops with a split-plot ANOVA with experimental site as the main plot and sowing date as the sub plot. For seedling crops, which consisted of two years of data, season was considered as a sub–sub plot. Seasonal growth rates were compared in the second season because of consistency of harvest dates. When differences occurred, indicated with a P-value ≤0.05, means were separated by Fishers least significant difference (l.s.d)

72

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

Fig. 1. Volumetric soil water content (SWC) at upper (䊉) and lower () limits of ‘Stamina 5’ lucerne water extraction measured to 2.3 m depth grown on a Wakanui silt loam (a) and a Lismore stony silt loam (b) soil at Lincoln University, Canterbury, New Zealand. Shaded areas and numbers represent plant available water content. Error bars represent l.s.d (P = 0.05).

at the 5% level. Regressions were calculated using the least-squares regression method. 3. Results 3.1. Shoot yield During the establishment season (i.e. from sowing to 30 June), delayed sowing from October until February reduced (P < 0.001) annual shoot yield in both sites (Figs. 2 a and 3 a ). Lucerne crops sown in October on the High PAWC site had a mean yield of 12.0 t DM ha−1 , or 85% more than the latest sown crops. At the Low PAWC site, yield was greatest when lucerne was sown from October to December, with a mean yield of 2.6 t DM ha−1 , or 80% more than when sown in February. In the regrowth season (1 July 2011–30 June 2012), sowing date in the previous establishment season affected yield at both sites. Crops which had been sown in spring (October–December) (P < 0.05) had a mean yield of 20 t DM ha−1 (Fig. 2b), and 7 t DM ha−1 (Fig. 3b), at the High and Low PAWC sites, respectively. Yield was reduced by 20–25% in the second season when sowing had been delayed beyond December, at both sites. 3.2. Seasonal growth patterns

3.3. Taproot plus crown biomass For sowing dates 1–5, taproot plus crown biomass at the end of the establishment season (June 2011) was influenced (P < 0.001) by sowing date. Maximum biomass was 5.0 t DM ha−1 , attained from crops sown from October to December on the High PAWC site (Fig. 5a) and 1.9 t DM ha−1 on the Low PAWC site (Fig. 5c), which was about 80% more than the February sown crops.

-1

Annual shoot yield (t DM ha )

Seasonal growth rates for sowing dates 1–5 in the first year were consistently ∼50 kg DM ha−1 day−1 for the crops at the High

PAWC site (Fig. 4a) and ∼15 kg DM ha−1 day−1 at the Low PAWC site (Fig. 4b). The exception was the October sown crop at the High PAWC with a growth rate of 85 kg DM ha−1 day−1 in the second regrowth cycle. In the second season, there was a seasonal pattern in shoot growth rate, irrespective of sowing date in the previous season. Overall, winter (June–July) regrowth cycles experienced close to zero growth (<5 kg DM ha−1 day−1 ). From spring (October) to autumn (April) shoot yield accumulated linearly with the mean highest growth rate in early summer (November) of 115 kg DM ha−1 day−1 and 45 kg DM ha−1 day−1 for the High and Low PAWC crops, respectively. The exception was the Low PAWC crops in the summer (January–February) regrowth cycle which experienced a mean growth rate of 10 kg DM ha−1 day−1 . In the second season, crops which had been sown early in the previous year had higher (P < 0.05) growth rates than the later sown crops. Specifically, growth rate for October 2010 sown crops were on average 10% higher in spring, and 30% higher in summer than the February 2011 sown crop in the 2011/2012 season.

25

a) Seedling season

b) Regrowth season

20 15 10 5 0

Oct

Nov

Dec

Jan

Feb

Oct

Nov

Dec

Jan

Feb

Sowing dates Fig. 2. Total annual shoot yield (t DM ha−1 ) of ‘Stamina 5’ dryland lucerne in the (a) seedling year and (b) subsequent regrowth year for crops sown in 2010/2011 (sowing dates 1–5, grey bars) and 2011/2012 (sowing dates 6–10, black bars) grown on a Wakanui silt loam soil (High PAWC) at Lincoln University, Canterbury, New Zealand. Error bars represent the standard error of the mean. Note: sowing dates are detailed in Table 1.

-1

Annual shoot yield (t DM ha )

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

73

10

b) Regrowth season

a) Seedling season 8 6 4 2 0

Oct

Nov

Dec

Jan

Feb

Oct

Nov

Dec

Jan

Feb

Sowing dates

150

b) Low PAWC

a) High PAWC

-1

Growth rate (kg DM ha day-1)

Fig. 3. Total annual shoot yield (t DM ha−1 ) of ‘Stamina 5’ dryland lucerne in the (a) seedling year and (b) subsequent regrowth year for crops sown in 2010/2011 (sowing dates 1–5, grey bars) and 2011/2012 (sowing dates 6–10, black bars) grown on a Lismore stony silt loam soil (Low PAWC) at Lincoln University, Canterbury, New Zealand. Error bars represent standard error of the mean. Note: sowing dates are detailed in Table 1.

100

50

0 Oct-10

Mar-11

Aug-11

Jan-12

Jun-12 Oct-10 Feb-11 Jun-11 Oct-11 Feb-12 Jun-12

Fig. 4. Seasonal growth rates (kg DM ha−1 day−1 ) of shoot dry matter (DM) of ‘Stamina 5’ dryland lucerne for individual growth cycles from sowing dates 1–5, sown in October (), November (), December (), January () and February (♦) grown on (a) high and (b) low plant available water capacity soil (PAWC) at Lincoln University, Canterbury, New Zealand. Error bars represent standard error of the mean. Dotted line represents end of establishment season. Note: sowing dates are detailed in Table 1.

10

-1

Taproot plus crown biomass (t DM ha )

8

a) High PAWC June 2011

b) High - June 2012

c) Low PAWC - June 2011

d) Low PAWC - June 2012

6 4 2 0 8 6 4 2 0 Oct

Nov

Dec

Jan

Feb

Oct

Nov

Dec

Jan

Feb

Sowing dates Fig. 5. Total annual taproot plus crown biomass (t DM ha−1 ) of ‘Stamina 5’ dryland lucerne for sowing dates 1–5 at the end of the establishment year (a,c) and year two (b,d) sown on five dates from October 2010 to February 2011 grown on a high (a,b) and low (c,d) plant available water capacity soil (PAWC) at Lincoln University, Canterbury, New Zealand. Error bars represent standard error of the mean. Note: sowing dates are detailed in Table 1.

74

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

12

a) Oct - High PAWC

f) Oct - Low PAWC

b) Nov - High PAWC

g) Nov - Low PAWC

c) Dec - High PAWC

h) Dec - Low PAWC

d) Jan - High PAWC

i) Jan - Low PAWC

e) Feb - High PAWC

j) Feb - Low PAWC

8

-1

Accumulated shoot and taproot plus crown biomass (t DM ha )

4 0 12 8 4 0 12 8 4 0 12 8 4 0 12 8 4 0 Oct-11

Jan-12

Apr-12

Oct-11

Jan-12

Apr-12

Jul-12

Fig. 6. Accumulated shoot (䊉) and taproot plus crown biomass () for ‘Stamina 5’ dryland lucerne for sowing dates 6–10 sown from October 2011 to February 2012 grown on a high (a–e) and low (f–j) plant available water capacity soil (PAWC) at Lincoln University, Canterbury, New Zealand. Note: Error bars represent standard error of the mean at the end of the establishment season.

3.4. Fractional partitioning of DM to taproot plus crown (Proot ) 3.4.1. Shoot and taproot plus crown biomass accumulation The accumulation of shoot and taproot plus crown biomass for sowing dates 6–10 is presented in Fig. 6. Root biomass increased up to a maximum of ∼5 t DM ha−1 in the early spring sown crops on the High PAWC site (Fig. 6a–c). The timing of the plateau in root biomass occurred at the second defoliation event, which differed with sowing date. This plateau occurred in late summer (21 February) for the earliest sown crop through to late autumn (11 April) for the December sown crop. Shoot DM continued to accumulate for these crops and was 2–3 times greater (P < 0.05) than root biomass by the end of the first year. All other crops had a root

0.6

Proot

At the end of the second season (June 2012), taproot plus crown biomass for crops on the High PAWC site was 6.7 t DM ha−1 , with the exception of the February sown crop which had a biomass of 5.7 t ha−1 , or 15% less (P < 0.05) (Fig. 5b). For the Low PAWC crops, greatest (P < 0.005) root biomass was observed from lucerne sown in October and November with 4.7 t DM ha−1 compared with 3.7 DM t ha−1 , or 22% less from the later sowing dates (Fig. 5d).

0.4

0.2

0.0 Sep-11

Dec-11

Mar-12

Jun-12

Date

Fig. 7. Fractional partitioning of DM to taproot plus crown (Proot ) for seedling (black symbols) and regrowth (white symbols) lucerne sowing dates 6–10 in relation to start date of growth for ‘Stamina 5’ dryland lucerne grown on a high () and low () plant available water capacity (PAWC) soil at Lincoln University, Canterbury, New Zealand. Error bar represents standard error of the mean.

biomass of <3 t DM ha−1 by June 2012, which were generally equal to accumulated shoot yields at that time. 3.4.2. Seasonal Proot patterns The fractional partitioning of DM to the taproots plus crown (Proot ) increased (P < 0.001) from 0.30 in spring (October) growth

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

Proot

b) Regrowth

a) Seedling

0.6

75

0.4

0.2 IPp DPp

0.0 0

12

14

16

18

0

12

14

16

18

Photoperiod (h) Fig. 8. Fractional partitioning of DM to taproot plus crown (Proot ) for seedling (a) and regrowth (b) lucerne in relation to increasing (IPp) and decreasing (DPp) photoperiod at the start of growth cycles for ‘Stamina 5’ dryland lucerne (sowing dates 6–10) grown on a high () and low () plant available water capacity (PAWC) soil at Lincoln University, Canterbury, New Zealand. For the seedling phase (a), regressions were only fitted to High PAWC data. For increasing photoperiod (—) y = 0.08x – 0.82, R2 = 0.97 and for decreasing photoperiod (—), the slope was not significantly different from zero (P = 0.05) and the mean Proot was 0.48. For regrowth cycles (b), High PAWC (—) y = 0.10x – 1.15, R2 = 0.75 and for Low PAWC (—) the slope was not significantly different from zero (P = 0.05) and the mean Proot was 0.45.

3.5. Reproductive development

Proot

0.6

0.4

0.2

0.0 0

2

4

6

8

Taproot plus crown biomass (t DM ha-1)

Fig. 9. Fractional partitioning of DM to taproot plus crown (Proot ) for seedling (black symbols) and regrowth (white symbols) lucerne in relation to taproot plus crown biomass at the start of growth cycle for ‘Stamina 5’ (sowing dates 6–10) dryland lucerne grown on a high () and low () plant available water capacity (PAWC) soil at Lincoln University, Canterbury, New Zealand. Solid line shows three part broken stick model fitted.

Irrespective of photoperiod at emergence, seedling crops displayed a constant thermal time requirement (P = 0.09) to reach the 50% visible bud stage, which differed (P < 0.05) with site (Fig. 10). On average, seedling crops at the High PAWC site required 705 ◦ Cd and at the Low PAWC site 780 ◦ Cd, or about 10% longer. In comparison, regrowth crops consistently displayed lower values which were affected (P < 0.05) by photoperiod at the start of the regrowth cycle. At the shortest photoperiod (11.3 h), the requirement was 555 ◦ Cd but declined (P < 0.05) at a rate of 75 ◦ Cd h−1 to 315 ◦ Cd at 14.2 h, which then became constant to the longest photoperiod of 16.7 h. 4. Discussion 4.1. Lucerne establishment phase

3.4.3. Proot in relation to taproot plus crown biomass Proot in relation to initial root biomass at the start of the growth cycle (Fig. 9) was described by fitting a broken stick model (R2 = 0.87). This indicated that Proot was 0.48 until a root biomass of 2.9 ± 0.28 t DM ha−1 , Proot then decreased at a rate of 0.0003 t DM−1 ha−1 to 0.05 when root biomass increased to 4.3 ± 0.32 t DM ha−1 . Proot then decreased to zero at a root biomass of 6.1 t DM ha−1 .

The lucerne establishment phase should be defined from sowing until crops have accumulated ∼5 t DM ha−1 of perennial biomass (Fig. 6). This phase extended beyond the seedling phase of lucerne (from sowing to first defoliation at flowering) and seemed to be regulated by demand of biomass to roots, rather than crop ontogeny. 1000 Seedling High PAWC Seedling Low PAWC Regrowth (Y1) High PAWC Regrowth (Y2) High PAWC Regrowth (Y2) Low PAWC

o

Thermal time to 50% buds visible ( Cd)

cycles to ∼0.50 in summer (January–February) to <0.10 in autumn for crops sown in the High PAWC site (Fig. 7). In contrast, Proot was 0.49, and did not differ (P = 0.60) amongst crops on the Low PAWC site. For the seedling phase, Proot for High PAWC lucerne increased (P < 0.05) from 0.32 in October to 0.51 in December. This increase in Proot showed a strong (R2 = 0.97) linear increase in relation to increasing photoperiod (IPp) at emergence (Fig. 8a). Proot increased by 0.08 h−1 when the Pp increased from 14.2 to 16.5. Proot was 0.53 (P = 0.06) for the seedling phase for crops on the Low PAWC. Proot in relation to Pp showed a moderate (R2 = 0.75) linear relationship for regrowth crops on the High PAWC site (Fig. 8b). Proot either increased or decreased by ∼0.10 for each hour depending on the direction of photoperiod change. Regrowth lucerne on the Low PAWC site did not respond to photoperiod (P = 0.60) and mean Proot was constant at 0.45.

800 600 400 200 0 0

11

12

13

14

15

16

17

Photoperiod at start of regrowth period (h)

Fig. 10. Thermal time (Tb = 1 ◦ C) requirement for appearance of visible flower buds for seedling and first (Y1; sowing dates 6–10) and second (Y2; sowing dates 1–5) year regrowth ‘Stamina 5’ dryland lucerne crops grown during a range of photoperiods in 2011/2012 season at Lincoln University, Canterbury, New Zealand. Error bar represents pooled standard error of the mean of seedling lucerne. Broken stick regression: (—) for regrowth crops y = −75.1x (x < 14.2) 314.8 (x > 14.2), R2 = 0.78.

76

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

During the establishment phase, lucerne prioritised the allocation of DM to the taproot plus crown of the plant, which explained the low rates of shoot growth in the first season, and also in the following season for crops that had not yet accumulated 5 t DM ha−1 . The optimal perennial biomass of 5 t DM ha−1 agrees with Teixeira et al. (2007a) who reported a similar value for lucerne defoliated at long (42 days) regrowth intervals. During this establishment phase, the fractional DM partitioned to the crown plus taproots (Proot ) was maintained at ∼0.50 until a minimum perennial biomass of ∼3 t DM ha−1 and then continued to accumulate to a maximum of ∼5 t DM ha−1 (Fig. 6). This priority of partitioning was also shown for the seedling phase by Khaiti and Lemaire (1992) and Thiebeau et al. (2011). As a consequence, it reduces the available supply of assimilates to shoots and explains the slower phenological development rates (phyllochron) for seedling lucerne relative to regrowth crops (Teixeira et al., 2011). In the present study, defoliation of seedling crops caused plant phenology to change and lucerne in the regrowth phase consistently reached reproductive development faster than the seedling phase (Fig. 10). This change in crop ontogeny was independent of DM partitioning and these regrowth crops continued to maintain a high Proot value until the optimal perennial biomass was attained. For example, when regrowth was initiated at a photoperiod of >14 h, the thermal time requirement to reach 50% visible bud stage for all regrowth crops was 315 ◦ Cd (Fig. 10). This is similar to the value reported by Teixeira et al. (2011) of 270 ◦ Cd for ‘Kaituna’ lucerne in the same environment. However, Proot values for these regrowth cycles ranged from close to zero to 0.50 (Fig. 8b). When establishing crops are added into the current partitioning framework (Teixeira et al., 2009), a unifying relationship between Proot and photoperiod for all crops was not evident in the present study (Fig. 8). Rather, the crop demand for perennial biomass offered insight into variability in partitioning during establishment. The establishment phase included the seedling phase and at least one further regrowth cycle. This pattern agrees with Khaiti and Lemaire (1992) who showed lucerne required three growth cycles to reach maximum perennial biomass. This means lucerne crops are likely to be less competitive aboveground for light in the establishment phase. Practically, it suggests pre and post emergence weed control, may be required to minimise competition for resources and maximise assimilate available to the perennial organs. Also, early defoliation of seedling crops to control weeds will interrupt the accumulation of biomass to the perennial organs, and most likely prolong the establishment phase as the limited root reserves are initially remobilised to support the subsequent regrowth cycle. 4.2. Seasonal Proot patterns Seedling lucerne did not display constant DM partitioning below ground which suggests a single Proot value was inappropriate for seedling growth. The Proot of seedling lucerne grown at the High PAWC site was described by the photoperiod at emergence by a similar seasonal pattern to that of regrowth lucerne (Teixeira et al., 2008) albeit at higher levels. However, environmental signals which regulate partitioning were over shadowed by water stress shown by the crops at the Low PAWC site, which displayed a conservative Proot of ∼0.50 (Fig. 7). This is consistent with Sheaffer et al. (1988), who reported generally absolute root biomass under water stress is less relative to fully watered crops, but increases proportionally to shoot biomass. For the crops sown on the Low PAWC site, growth rates were a third of that grown on the High PAWC site (Fig. 4). Water limited growth reduced the amount of assimilate available to be partitioned below ground, which meant the perennial biomass did not exceed 3 t DM ha−1 until the second season (Fig. 5) and Proot were maintained at ∼0.50 for longer. It seems apparent that these crops may need to be treated as estab-

lishing crops beyond the first defoliation event and the period of establishment is likely to be regulated by environmental conditions which influence shoot growth. For rainfed lucerne water supply largely determines shoot growth (Moot et al., 2008), and therefore, the seasonal rainfall distribution will influence the length of the establishment phase. Proot of regrowth lucerne in the second year most likely responded to seasonal signals, although late sown crops were still in the establishment phase. This was most apparent in the crops at the High PAWC site, where delayed sowing from October to February reduced root biomass at the end of the first year from 5.3 to 1.3 t DM ha−1 (Fig. 5a). In the second season, crops preferentially partitioned DM to the perennial organs, and by the end of the second season, all crops had a biomass of >5 t DM ha−1 (Fig. 5b). This most likely reduced the assimilate supply for the expansion of leaf area (Teixeira et al., 2007c) and as a consequence, shoot growth rates for these crops were consistently reduced by ∼15% in spring, and up to ∼30% in summer (Fig. 4a). This implies environmental factors dominated Proot , which agrees with Teixeira et al. (2008) who showed partitioning in lucerne responded to photoperiod, irrespective of absolute perennial biomass. However, agronomic management which reduces the perennial biomass below the minimum threshold, shown in the present study to be ∼3 t DM ha−1 , such as delayed sowing, inadequate nitrogen supply (Avice et al., 2003; Teixeira et al., 2008) or short defoliation intervals (Teixeira et al., 2007a) will most likely regulate this perennial biomass demand. This means delaying the first spring gazing is unlikely to build up sufficient root reserves for late sown crops. This reinforces the recommendation to graze at the appearance of open flowers for at least one regrowth cycle during the periods of decreasing photoperiod to maximise recharge of reserves (Moot et al., 2003). 4.3. PAWC influenced seedling reproductive development The difference in thermal time requirement to reach reproductive development for seedling crops between sites (Fig. 10) can be attributed to differences in leaf appearance rates (data not shown). Once mainstem leaf production was initiated, which occurred at 124 ± 26 ◦ Cd after emergence, 11.6 ± 0.3 leaves were expanded (P = 0.24) before crops reached the 50% visible bud stage. The mean rate of leaf appearance (phyllochron) was 45 and 58 ◦ Cd (P < 0.001) for seedling crops at the High and Low PAWC site, respectively. Therefore, when the appearance of mainstem leaves was delayed, time to reach reproductive development extended. Most likely, water stress reduced the extension and unfolding of individual leaves (Brown and Tanner, 1983), which decreased the leaf appearance rate, which has been shown for severe water limited lucerne (Brown et al., 2009). In the present study, the additional thermal time to reproductive development amongst seedling crops did not show a consistent correlation (R2 = 0.43) to greater Proot , which further supports the hypothesis that crop ontogeny is independent of assimilate partitioning during the establishment phase. 5. Conclusions The lucerne establishment phase was defined by the time from sowing until crops reached a maximum perennial biomass accumulation of ∼5 t DM ha−1 . Establishment extended beyond the seedling phase and was independent of crop ontogeny. To maximise shoot yield, lucerne should be sown in spring to allow sufficient time to build perennial biomass to optimum levels. This research has provided insight into the sink demand of the perennial organs during the establishment phase which may be used to develop agronomic management practices and incorporated into current partitioning frameworks for crop modelling, which is cur-

R.E. Sim et al. / Europ. J. Agronomy 68 (2015) 69–77

rently largely dependent on crop ontogeny and environmental influences. Acknowledgements The William Machin Trust funded Richard Sim’s Ph.D. Research, and Lincoln University and Beef + Lamb NZ assisted financially to allow preparation of this manuscript. References Avice, J.C., Dily, F.L., Goulas, E., Noquet, C., Meuriot, F., Volenec, J.J., Cunningham, S.M., Sors, T.G., Dhont, C., Castonguay, Y., Nadeau, P., Belanger, G., Chalifour, F.P., Ourry, A., 2003. Vegetative storage proteins in overwintering storage organs of forage legumes: roles and regulation. Can. J. Bot. 81, 1198–1212. Brown, P.W., Tanner, C.B., 1983. Alfalfa stem and leaf growth during water stress. Agron. J. 75, 799–804. Brown, H.E., Moot, D.J., Fletcher, A.L., Jamieson, P.D., 2009. A framework for quantifying water extraction and water stress responses of perennial lucerne. Crop Pasture Sci. 60, 785–794. Cox, J.E., 1978. Soils and agriculture of Paparua County Canterbury Soil Bureau Bulletin No.34. DSIR, Wellington, New Zealand. Di, H.J., Cameron, K.C., 2002. Nitrate leaching and pasture production from different nitrogen sources on a shallow stoney soil under flood-irrigation dairy pasture. Aust. J. Agric. Res. 40, 317–334. Fick, G.W., Holt, D.A., Lugg, D.G., et al., 1988. Environmental physiology and crop growth. In: Hanson, A.A. (Ed.), Alfalfa and Alfalfa Improvement. American Society of Agronomy, Madison, USA, pp. 163–194. Fornasier, F., Pecceti, L., Piano, E., 2003. Variation in crown and root organic reserves among lucerne genotypes of different morphology and flower colour. J. Agron. Crop Sci. 189, 63–70. Goodspeed, M.J., 1975. Computer routines for solar position, daylenght and related quantities. CSIRO Australia Division of Land Use Research Technical Memorandum 75/11. Jones, C.A., Kiniry, J.R., 1986. CERES-Maize: A Simulation Model of Maize Growth and Development. Texas A&M University Press, College Station, Texas, USA. Justes, E., Thiebeau, P., Avice, J., Lemaire, G., Volenec, J., Ourry, A., 2002. Influence of summer sowing dates, N fertilization and irrigation on autumn VSP accumulation and dynamics of spring regrowth in alfalfa (Medicago sativa L.). J. Exp. Bot. 53, 111–121. Khaiti, M., Lemaire, G., 1992. Dynamics of shoot and root growth of lucerne after seeding and after cutting. Eur. J. Agron. 1, 241–247. Khumalo, Q., Moot, D.J., Wigley, K., 2012. Yield, final population and emergence of seed treated lucerne (Medicago sativa L.) sown on five dates. In: Harris, C. (Ed.), Proceedings of the Inaugural Australian Legume Symposium, William Angliss Conference Centre, 58–60. Lemaire, G., Khaiti, M., Onillon, B., Allirand, J.M., Chartier, M., Gosse, G., 1992. Dynamics of accumulation and partitioning of N in leaves, stems and roots of lucerne (Medicago sativa L.) in a dense canopy. Ann. Bot. 70, 429–435. Mills, A., 2007. Understanding Constraints to Cocksfoot (Dactylis glomerata L.) Based Pasture Production. PhD Thesis, Lincoln University, Canterbury, New Zealand.

77

Moot, D.J., Robertson, M.J., Pollock, K.M., 2001. Validation of the APSIM-Lucerne model for phenological development in a cool-temperate climate. Proceedings of the 10th Australian Agronomy Conference, 1–5. Moot, D.J., Brown, H.E., Teixeira, E.I., Pollock, K.M., 2003. Crop growth and development affect seasonal priorities for lucerne management. In: Moot, D.J. (Ed.), Legumes for Dryland Pastures. New Zealand Grassland Association, Lincoln University, Christchurch, New Zealand, pp. 201–208. Moot, D.J., Brown, H.E., Pollock, K., Mills, A., 2008. Yield and water use of temperate pastures in summer dry environments. Proc. N. Z. Grassland Assoc. 70, 51–57. Moot, D.J., Pollock, K.M., Lewis, B., 2012. Plant population, yield and water use of lucerne sown in autumn at four sowing rates. Proc. N. Z. Grassland Assoc. 74, 97–102. Morton, J., Roberts, A., 1999. Fertilizer Use on New Zealand Sheep and Beef Farms. New Zealand Fertiliser Manufacturer’s Association, Auckland, pp. 36. Robertson, M.J., Carberry, P.S., Huth, N.I., Turpin, J.E., Probert, M.E., Poultron, P.L., Bell, M.J., Wright, G.C., Yeates, S.J., Brinsmead, R.B., 2002. Simulation of growth and development of diverse legumes species in APSIM. Aust. J. Agric. Res. 53, 429–446. Sheaffer, C.C., Tanner, C.B., Kirkham, M.B., et al., 1988. Alfalfa water relations and irrigation. In: Hanson, A.A. (Ed.), Alfalfa and Alfalfa Improvement. American Society of Agronomy, Madison, WI, pp. 373–409. Sim, R.E., 2014. Water extraction and use of seedling and established dryland lucerne crops. In: PhD Thesis. Lincoln University, Canterbury, New Zealand. Teixeira, E.I., Moot, D.J., Mickelbart, M.V., 2007a. Seasonal patterns of root C and N reserves of lucerne crops (Medicago sativa L.) grown in a temperate climate were affected by defoliation regime. Eur. J. Agron. 26, 10–20. Teixeira, E.I., Moot, D.J., Brown, H.E., Fletcher, A.L., 2007b. The dynamics of lucerne (Medicago sativa L.) yield components in response to defoliation frequency. Eur. J. Agron. 26, 394–400. Teixeira, E.I., Moot, D.J., Pollock, K.J., Brown, H.E., 2007c. How does defoliation management affect yield, canopy forming processes and light interception in lucerne (Medicago sativa L.) crops? Eur. J. Agron. 27, 154–164. Teixeira, E.I., Moot, D.J., Brown, H.E., 2008. Defoliation frequency and season affected radiation use efficiency and dry matter partitioning to roots of lucerne (Medicago sativa L.) crops. Eur. J. Agron. 28, 103–111. Teixeira, E.I., Moot, D.J., Brown, H.E., 2009. Modelling seasonality of dry matter partitioning and root maintenance respiration in lucerne (Medicago sativa L.) crops. Crop Pasture Sci. 60, 778–784. Teixeira, E.I., Brown, H.E., Moot, D.J., Meenken, E.D., 2011. Growth and phenological development patterns differ between seedling and regrowth lucerne crops (Medicago sativa L.). Eur. J. Agron. 35, 103–111. Thiebeau, P., Beaudoin, N., Justes, E., Allirand, J., Lemaire, G., 2011. Radiation use efficiency and shoot:root dry matter partitioning in seedling growths and regrowth crops of lucerne (Medicago sativa L.) after spring and autumn sowings. Eur. J. Agron. 35, 255–268. Volenec, J.J., Ourry, A., Joern, B.C., 1996. A role for nitrogen reserves in forage regrowth and stress tolerance. Physiol. Plant. 97, 185–193. Webb, T.H., 2003. Identification of functional horizons to predict physical properties for soils from alluvium in Canterbury, New Zealand. Aust. J. Soil Res. 41, 1005–1019. Wigley, K., Moot, D.J., Khumalo, Q., Mills, A., 2012. Establishment of lucerne (Medicago sativa) sown on five dates with four inoculation treatments. Proc. N. Z. Grassland Assoc., 91–96.