Capture and use of solar radiation, water, and nitrogen by sugar beet (Beta vulgaris L.)

Abstract

Sugar beet is spring-sown for sugar production in most sugar beet-growing countries. It is grown as a vegetative crop and it accumulates yield (sugar) from very early in its growth cycle. As long as the sugar beet plants do not flower, the sugar accumulation period is indefinite and yield continues to increase. This paper reviews the success of the sugar beet crop in capturing and using solar radiation, water and mineral nitrogen resources. The prospects for improved resource capture and therefore increased sugar yield are also considered, particularly the potential to increase solar radiation interception in the future by sowing the crop in the autumn.

Autumn beet, root distribution, winter beet

Introduction

A meeting concerned with resource capture by crops was last held at Nottingham University 15 years ago (Scott et al. 1994). At that time there was a major concern that the value of European crops would fall and that crop growth needed to use resources efficiently if farming was to be an attractive and prosperous activity. Since then crop prices have fallen, risen again dramatically and then fallen just as quickly. At the same time, prices of energy and fertilizers have changed like the profile of a rollercoaster, and at a similar pace. This harsh economic climate, coupled with the need for agriculture to have lesser impacts on the natural environment means that efficient use of resources by crops remains a top priority. In this paper, we examine how the use of the light, water and nitrogen resources has changed in recent times, and consider the prospects for improvements in the future. As in the paper from 15 years ago, we have concentrated on sugar beet, partly because it is a simple crop that, if grown properly, does not flower and accumulates yield (sugar) from very early in its growth cycle. Growth is a product of the summed daily increments of intercepted radiation and the radiation use efficiency; thus, each day the resources captured by the plant make a small but important contribution to final yield. The simple relationships between the amounts of resources captured and the productivity of the crop can be described using a mechanistic growth model, which can then be used to predict the impact of various factors on yield.

Capture and use of solar radiation

More than 50 years ago, it was recognized that the principal failing of the sugar beet crop was its poor ability to capture solar radiation—the canopy of the crop developed so late in the summer that much of the available radiation was not intercepted (Watson, 1956). However, plant breeders and agronomists have made efforts to improve radiation capture and yields in the UK, and elsewhere, have risen. How much of this improvement has been due to increased radiation capture and how has this been achieved? Annual average yields of sugar for the UK crop since 1976 are plotted in Fig. 1. In commercial crops, yield has increased by an average of 110 kg ha⁻¹ year⁻¹; in official trials used to assess varieties for recommendation and use, the increase was 202 kg ha⁻¹ year⁻¹. Initiatives to improve seed quality, bolting resistance and seedbed preparation have enabled earlier sowing dates so that the foliage canopy development is better synchronized with the availability of solar radiation. The date by which 50% of the UK crop was sown varies greatly each year (Fig. 2), mostly due to the timing of periods in spring when the soil was dry enough for seed drills to operate effectively. In recent years, most of the UK crop was sown in March.

Fig. 1.

Average sugar yield of sugar beet in England and in the National Institute of Agricultural Botany (NIAB) variety trials each year: 1976–2006.

Fig. 2.

The date when 50% of the UK sugar beet crop was sown each year: 1976–2006.

The 50% sown date and the weather data each year (measured at Broom's Barn Research Centre in Suffolk, UK) was used to run a crop growth model (Qi et al., 2005) to simulate the sugar yield of crops grown without irrigation in sandy loam soil. These simulations have shown an average yield increase of 118 kg ha⁻¹ year⁻¹, which is similar to the observed yield increase. Importantly, none of the simulated yield increase was due to changes in variety, because the model operates using a ‘generic’ variety without any variety-specific inputs. The annual deviations from the trend line tend to have the same sign as the deviations from the trend in the official variety trials, which indicates that the model accurately predicts actual yields (Jaggard et al., 2007). The trend for increased yields was partly due to earlier sowing, but most of the increase was due to warmer summers (Fig. 3). The combination of earlier sowing and more rapid expansion of leaves in warmer temperatures increased the amount of radiation captured by an earlier closure of the crop canopy. The expansion of the canopy of the sugar beet crop, beginning from the emergence of small seedlings (with <1 cm² leaf area per plant) to complete ground coverage is a temperature-dependent process. Therefore, any warming of the weather in spring advances canopy cover and increases solar radiation interception. There have also been large variations in summer rainfall over this 30-year period, but there is no indication that the summers have tended to become wetter or drier at Broom's Barn (Fig. 3). Changes to the climate and changes to the time of sowing have been responsible for a yield increase of 138 kg ha⁻¹ year⁻¹, or 68% of the total increase observed in variety trials (Table 1). The remainder of the yield increase can be attributed to several other factors (Jaggard et al., 2007).

Table 1.

Allocation of annual rates of sugar yield increases measured in variety trials according to growth simulations using the sugar beet growth model during1976–2006

	Annual sugar increase (kg ha−1 year−1)	Percentage (%)
Total increase	202	100
Earlier sowing	20	10
Climate change	118	58
Residual (plant breeding, agronomy, CO2)	64	32

Fig. 3.

Average air temperature and total rainfall measured at 18 weather stations in eastern England where the sugar beet crop is grown from 1 March to 31 October each year: 1976–2006; the vertical bars represent the range amongst the 18 stations.

In addition to increased radiation interception, another potential avenue of improvement is through improved radiation use efficiency (RUE). In England the values for season-long RUE in sugar beet were between 1.4 and 1.7 g MJ⁻¹ throughout the 1980s (Werker and Jaggard, 1998) and varied due to water stress and the age of the canopy. Young, stress-free crops had an average of 1.76 g MJ⁻¹, which compares very favourably with rates recorded in sugar cane, a C₄ species that also stores sucrose (Sinclair and Muchow, 1999). There is no evidence for an improvement in RUE since then. Because water deficit can decrease RUE (Ober et al., 2004), the avoidance of drought stress, either through irrigation or through the use of improved varieties, are two possible solutions. Comparison of a wide range of sugar beet genotypes under managed drought conditions in the field has shown that there is significant genotypic variation for radiation use efficiency and drought tolerance (Ober et al., 2004). This information helps equip breeders to make selections for varieties that are better able to utilize water and radiation.

Capture of water

It is generally believed that current varieties yield better than much older varieties, and therefore that resource capture and utilization have also improved. It is often difficult to test these assumptions. Every year at Broom's Barn two ‘barometer’ beet crops are grown using the best agronomic practices, one with irrigation and the other without. The barometer crop of cv. Regina grown with irrigation in 1982 set a local record for sugar production that stood until 2006, when the irrigated crop (cv. Anemona) set a new record of over 18.0 t ha⁻¹ of sugar (Table 2). The canopy expansion period (April–June) was 0.3 °C warmer in 2006 than in 1982, the summer of 2006 was slightly brighter than 1982, and the irrigated crop intercepted a slightly larger fraction of the summer radiation (an extra 1.6%). Although different varieties were grown in 1982 and 2006, the two irrigated crops had similar season-long radiation use efficiencies, suggesting that this characteristic has not improved through breeding in the intervening 24 years. However, the efficiencies of both crops were larger than the corresponding rain-fed crops. The irrigated crops consumed more than 350 mm of water between June and September, 100 mm more than their rain-fed counterparts. Therefore, the crop in 2006 had not managed to exploit the soil water and summer rainfall any better than the crop in 1982 (Table 2).

Table 2.

Characteristics of high yield sugar beet crops grown at Broom's Barn in 1982 and 2006

Variable	1982		2006
	Irrigated	Rain-fed	Irrigated	Rain-fed
Total dry matter (t ha−1)	27.3	22.4	28.6	22.7
Sugar (t ha−1)	15.4	13.1	18.4	13.7
Sugar concentration (%)	17.7	17.8	18.0	17.5
Radiation receipt (MJ m−2)	2889	2889	3146	3146
Intercepted radiation. (MJ m−2)	1866	1794	2083	1865
Radiation intercepted (%)	64.6	62.1	66.2	59.3
RUE (g MJ−1)	1.46	1.26	1.37	1.22
Water consumption June–September (mm)	354	250	385	259

Root development and water uptake was studied in detail in the 1982 crop (Brown and Biscoe, 1985) and such comprehensive measurements have not been made since: there is no alternative but to rely on the old data. The development of the fibrous root system in 1982 is shown in Fig. 4. Although the roots quickly extended down to a depth of 1 m, they did not get beyond this level until late summer. These deeper layers are occupied relatively late in relation to the periods of intense water demand so that little use is made of the water stored there. In fact, observations of root systems in situ have found that, frequently, many of the fibrous roots deep in the subsoil pass through old, well-established pores. This means that the root system may be clumped and not capable of efficiently using the soil's water reserve. In 1982, the rain-fed crop extracted almost all the available water from the uppermost 60 cm of soil (Table 3), but available water was still present in the lower layers while the crop was responding strongly to irrigation. Clearly there is an opportunity for beet crops to be improved in order to increase the capture of water from the lower parts of the soil profile. There is evidence of significant genotypic differences in root activity in deep soil layers measured under droughted conditions (Ober et al., 2005a, b). Although currently it is not practical to make selections for such root traits in sugar beet breeding programmes, knowledge of the extent of genetic variation and identification of superior genotypes is useful information.

Table 3.

Water use by rain-fed sugar beet crop grown in a sandy loam soil at Broom's Barn in 1982

Soil depth (cm)	Max. soil water used (mm)	Available water (mm)	Available water used (%)	Water removed (mm)
0–30	42.0	42.5	99	199.0
30–60	29.5	31.5	94	29.5
60–100	18.5	36.0	51	18.5
100–155	3.0	61.5	4	5.0
0–155	93.0	171.5	54	252.0

Fig. 4.

The distribution of root length density (cm cm^-3 of soil) on different dates in a sugar beet crop grown at Broom's Barn without irrigation in 1982 (After Brown and Biscoe, 1985).

Capture of nitrogen

The principal role of fertilizer N in the sugar beet crop is to stimulate the production of a foliage canopy that will allow approximately 90% of the solar radiation to be intercepted as soon after sowing as possible (Malnou et al., 2006): it does not improve RUE in the latter half of the growing season, when the canopy is ageing (Malnou et al., 2008). Mineral nitrogen in the soil and in the crop was assessed in experiments to estimate the optimum dose of N fertilizer in 33 situations between 1980 and 2001 at Broom's Barn. These data are a subset of the N responses analysed by Jaggard et al. (2009). Average values are shown in Table 4, for crops that will typically have produced 20–25 t ha⁻¹ of dry matter. Following a cereal crop, the top 1 m of sandy loam soil in spring contains about 60 kg N ha⁻¹ (Allison et al., 1996). When this is supplemented with 120 kg N ha⁻¹ from fertilizer, the beet crop will reach its maximum N uptake of 200–250 kg N ha⁻¹ by late August or early September. In these soils the residual mineral N in the top metre when the beet are harvested is seldom more than 30 kg ha⁻¹. Thus, net N mineralization is approximately 60 kg ha⁻¹ (Table 4) and the beet crop apparently captures approximately 80% of the available N. In reality, it captures a smaller percentage because mineral N is continuously being converted to organic forms by the soil microflora. While further gains are possible, compared with other crops, sugar beet is already efficient at capturing and utilizing soil N.

Table 4.

Average values of nitrogen supply, uptake and residue in the soil at harvest for sugar beet crops grown in studies of 33 nitrogen responses at Broom's Barn between 1980 and 2001

Source/fate	kg ha−1
Soil at sowing	58 (26.4)
Fertilizer	120
Net mineralization	59 (33.6)
Crop uptake	201 (39.6)
Residue at harvest	24 (15.8)

The data are all from plots receiving 120 kg N ha⁻¹ of fertilizer, growing in a sandy loam soil which contained no more than 2% organic matter. Values in parentheses are the standard deviations.

Future prospects for resource capture

The sugar beet growth model (Qi et al., 2005) can be used to assess the likely yield effects of changes to varieties and/or crop agronomy to increase the capture of solar radiation or soil water. In this paper, we have concentrated on improvements in the capture of solar radiation. Clearly there are two possible routes: to increase the rate at which the foliage expands per unit of thermal time or to modify the agronomy so that seedlings emerge from the soil earlier.

Today's varieties take approximately 1050 °C days (°Cd) from sowing to achieve 90% foliage cover. If seedlings could be produced with the capacity to reach the same stage in 850 °Cd then this would have increased sugar yield in England in 2006 by about 1 t ha⁻¹ (Table 5). To make seedlings emerge earlier, the choice is between making a reduction in the thermal time from sowing to emergence, or to sow earlier. Already commercial beet seeds are primed prior to sowing to accelerate germination, so the scope for further thermal time reductions is small. A reduction from the baseline value of 120 °Cd for 50% of the seedlings to emerge to 90 °Cd has just been introduced (Burks, 2008): a further reduction to 80 °Cd should be possible, and this should increase sugar yield by about 0.3 t ha⁻¹ compared with the seed used in England in 2007.

Table 5.

Simulated sugar yields of beet having different periods of thermal time from sowing to 90% foliage cover

Scenario (°Cd)	Sugar yield (t ha−1)
1050 (baseline)	13.57
950	14.06
850	14.42
750	14.73
650	15.14
550	15.46

The simulations are all based on a common sowing date (22 March 2006) and harvest on 7 December 2006.

Within a population of seedlings of a given variety, there is a great deal of variation in the initial rate of leaf expansion. This is because sugar beet varieties are heterogeneous populations typically produced by crossing parental lines that are not inbred, unlike maize for example (Bosemark, 2006). Figure 5 shows 42 seedlings from the same seedlot of a commercial variety (Hilleshog's Celt). Each seed was sown carefully by hand at precisely the same depth and all produced a seedling, so the seed lot was of the highest quality. The variation in leaf area among the seedlings was huge, despite grading the seed carefully to minimize variation in seed size. The very small seedlings emerged through the soil surface several days after the bulk of the population, but most of the variation could not be associated with emergence date. Milford and Riley (1980) showed that, among some older varieties, improvements in the rate of leaf growth at cool, spring temperatures had been made, but most of the variation was within varieties. Despite the observation that significant genotypic diversity for relative leaf expansion rates exists (Ober and Luterbacher, 2002), little progress in selecting for rapid rates of leaf expansion has been made so far (Bosemark, 2006). Differences in seedling vigour reflect the heterogeneous nature of sugar beet hybrid populations, but currently there is no reliable way to predict whether an individual seed will produce a large or a small seedling. As yet, no experiment has shown that a population composed entirely of the large seedlings would produce a significant increase in yield, but it is certainly the case that anything that decreases seedling size in early summer decreases yield. It is therefore reasonable to assume that potential yield is lost in non-uniform stands of seedlings because space is inefficiently occupied by ‘weaker’ seedlings, which cannot be fully recovered by the compensatory growth of the more vigorous seedlings. If varieties can be created where the larger seedlings make up more of the population then a yield increase of 1 t ha⁻¹ should be achievable.

Fig. 5.

Photograph of hand-sown sugar beet seedlings from a commercial seed lot, showing the variation in seedling leaf area. This variation represents the potential to increase the rate of leaf growth in varieties in future.

Capture of more radiation earlier in the season could be accomplished by earlier sowing. In theory, the sowing date could be advanced significantly by sowing the previous autumn. Experiments in Germany have shown that seeds sown in late autumn can give rise to plants that survive the winter and then go on to intercept 40% of the solar radiation in April and 73% in May, whereas conventional sowings intercept almost none of this energy (Anon, 2008). These values are similar to simulations made for crops in England (Fig. 6). However, the autumn-sown plants become vernalized over winter and bolt, making the crop worthless. Therefore, plant breeders are working to produce varieties that are extremely resistant to bolting and flowering so that beet can be sown in autumn. In eastern England, seeds sown in August or early September almost always survive the winter, so frost hardiness is not a serious issue for our climate, but there would be many disease control issues to be solved (Jaggard and Werker, 1999). Nevertheless, plant breeders are now more confident that, with the application of modern biotechnology, they can overcome these disease problems. Therefore, making the conservative assumptions listed in Table 6 (Jaggard and Werker, 1999) it was estimated that autumn-sown beet in England in 2006 would have increased sugar yield by more than one-third due to extra intercepted radiation (Table 7; Fig. 6). One of the assumptions was that the rooting and water extraction would be no deeper than in spring-sown crops, whose yield in eastern England is often restricted by the availability of water. If the longer growing season provided roots with the opportunity to penetrate the soil more deeply, the water restriction might be less acute and the yield increase even larger.

Table 6.

Assumptions made when simulating the growth of autumn-sown and spring-sown sugar beet crops using the model of Qi et al. (2005)

	Autumn-sown crop	Spring-sown crop
Sowing date	1 September	Mean date of the first 10 soil work days in spring
Foliage and root growth cease	First air frost of below –1 °C	–
Overwinter foliage cover	4%	–
Overwinter root depth	Depth at time of first air frost	–
Start of spring growth	Date of sowing of spring crop	Date of sowing
Maximum rooting depth	150 cm	150 cm
Harvest date	31 October	31 October

Table 7.

The simulated changes in yield and radiation interception of rain-fed beet crops sown in autumn on 1 September 2005 (autumn-sown) or in spring on 22 March 2006 (spring-sown) and harvested at the end of October 2006

Variable	Spring-sown	Autumn-sown	Increase
			Quantity	%
Total dry matter (t ha−1)	23.6	30.3	6.7	28.4
Sugar yield (t ha−1)	13.6	18.8	5.3	38.2
Intercepted radiation (MJ m−2)	1994	2265	271	13.6

Fig. 6.

The monthly radiation receipt and the radiation intercepted in 2006 by simulated sugar beet canopies of crops sown in a sandy loam soil at Broom's Barn on 1 September 2005 (autumn-sown beet) and on 22 March 2006 (spring-sown beet).

Sugar beet pathologists, weed control specialists and soil scientists are working to reduce the harmful impacts of pests, diseases and weeds and will seek to improve the condition of the soil for the growth of roots and the extraction of nutrients. These changes will tend to reduce the gap between the potential yield of the crop and the yield that is delivered to the sugar processing factory, but what are the prospects for improving the potential yield? In this regard, the sugar beet breeder will need to balance very carefully the allocation of dry matter to the various plant organs. Early in growth, the ideal plant is allocated more dry matter to produce a larger canopy which intercepts more sunlight and produces more photosynthates. A small portion of these extra photosynthates will need to be allocated to the growth of fibrous roots deep in the soil profile so that the crop can acquire the water it needs to grow efficiently. Late in the summer the breeder needs to select genotypes which allocate much of their new dry matter to stored sugar and not foliage growth.

Conclusions

There is evidence that, in recent decades, there have been improvements in the capture of the solar radiation resource by sugar beet crops in the UK, and yield has increased as a consequence. In part, this has been achieved by attention to seed quality and seedbed preparation techniques so that sowing could be safely advanced into the middle of March. However, the greatest improvement has come about as a result of warmer spring and summer weather in the last 30 years, which has accelerated the rate of canopy development. Beet crops still fail to intercept more than 30% of the solar radiation received between the beginning of April and the end of October (Fig. 6), so there is still much scope for improvement because total solar radiation receipts are directly related to yield gain. However, any increase in the amount of radiation intercepted will inevitably mean that the potential water consumption by the crop will increase (because carbon fixation and water loss are tied energetically). Unfortunately, a comparison of old and new varieties suggests that the capture of available water from the reserve deep in the soil has not increased over the years of varietal improvement. While the uppermost 60 cm of soil can be completely emptied of available water by beet roots, little more that 5% of the water is extracted from between 100 cm and 150 cm depth.

There is considerable potential to increase solar radiation interception in the future, and to do it at a time of the year when the demand for water is not at its peak. Plant breeders are working to create beet varieties that can be sown successfully in autumn for harvest one year later. The basic principles of resource capture can be illustrated using sugar beet as a simple model, and with additional considerations (e.g. the phenology of flowering), these concepts can be applied to other crops. If in another 15 years there is another conference on resource capture, the hope is that these fundamentals of crop science won't be merely rediscovered, but that they are built upon and used for improvements in genetics, agronomy and farm management.

Much of the paper is based on work funded by the British Beet Growers Organization (BBRO). Broom's Barn Research Centre is part of Rothamsted Research, which receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.