Agronomy Journal 93:290-298 (2001)
© 2001 American Society of Agronomy
SYMPOSIUM PAPERS
Increasing Water Use and Water Use Efficiency in Dryland Wheat
J.F. Angus and
A.F. van Herwaarden
CSIRO Plant Industry, GPO Box 1600, Canberra, 2601, Australia
Corresponding author (jfa{at}pi.csiro.au)
 |
ABSTRACT
|
|---|
Water use efficiency (WUE), the ratio of grain yield to crop water use, provides a simple means of assessing whether yield is limited by water supply or other factors. Based on this assessment, yields of commercial dryland wheat (Triticum aestivum L.) crops in southeastern Australia are usually not limited by water. Transpiration efficiency (TE), the ratio of yield to transpiration, is relatively stable for well-managed crops, but the amount of water used is strongly affected by crop management. In a review of 13 comparisons of water use and wheat yield, providing optimum N fertilizer or suppressing cryptic root diseases with break crops increased water use by 23 mm and yields by 378 kg ha-1, equivalent to 10% of the control yields. The additional soil water was extracted to levels of water potential as low as -5 MPa. A possible means of increasing yield potential of dryland crops is to manage transpiration so that relatively more water is used during the vegetative phase when vapor pressure deficit (VPD) is low, and hence TE is high. However, based on budgets of soil water and soluble carbohydrates stored in the vegetative organs and available for retranslocation, this option provides lower TE than conserving soil water for transpiration until grain filling when assimilates are directed to grain. Increasing the proportion of water transpired during the vegetative phase with N fertilizer can lead to particularly inefficient water use because increasing N status generally reduces the soluble carbohydrate reserves available for retranslocation to grain.
Abbreviations: Es, soil evaporation • ET, evapotranspiration • TE, transpiration efficiency • VPD, vapor pressure deficit • WLY, water-limited yield • WUE, water use efficiency
|
INTRODUCTION
|
|---|
WATER USE EFFICIENCY continues to interest plant and soil scientists 17 yr after publication of the ASA monograph edited by Taylor et al. (1983), despite the review in that book by Tanner and Sinclair (1983), who asserted that much of the research over the previous 75 yr had been "re-search" of earlier results and conclusions. Since 1979, over 2100 papers referring to WUE have been published, according to the CABI database. This paper reviews the contribution made by the WUE concept to commercial wheat production in southeastern Australia and two new aspects related to water use and WUE that are important in Australia and may have applications in other dryland grain regions.
Research on WUE is an attempt to simplify the complex mechanisms relating water use and yield. A concise expression of the relationship is presented in Eq. [1]:
 | (1) |
where 
W is growth (kg ha-1), ET is evapotranspiration (mm), Es is soil evaporation (mm), e* is saturated vapor pressure (kPa), and e is actual vapor pressure (kPa). The empirically determined crop-specific constant k has units of kPa mm-1. The value of ET can be calculated using Eq. [2]:
 | (2) |
where R is rainfall during a defined period, and SW represents the soil water content of the root zone at the start and end of the period. Other terms in the water balance, surface runoff, and drainage are usually negligible on flat land in semiarid and subhumid environments. The units for all of the terms in Eq. [2] are millimeters of water. For ET over the whole life cycle of the crop, the start is normally the date of sowing, and the end is the date of maturity. The antecedents of the WUE approach date back 90 yr to the research on transpiration ratio by Briggs and Shantz (1916) and others in the Great Plains of the USA. They showed that the yield of plants was linearly related to transpiration. Even earlier research on WUE was reviewed by Richardson (1923).
There were two key steps in developing the theory from the transpiration ratio to the summary expressed in Eq. [1] and [2]. One was to separate transpiration from Es in the field. Hanks et al. (1969) provided an approximation from a graph of dry matter production plotted against ET that was extrapolated to Es on the abscissa. The term ET - Es in Eq. [1] represents transpiration. The second step was to resolve the observations by Briggs and Shantz (1916), de Wit (1958), and others that TE was low when the evaporative demand was high. Bierhuizen and Slatyer (1965) showed that TE was linearly related to the VPD, which is defined as the difference between saturated vapor pressure, e*, and actual vapor pressure, e, at the same temperature. The VPD is proposed as the most appropriate field measure of the evaporative demand because it approximates the gradient in vapor concentration between the saturated leaf mesophyll and the atmosphere.
Because the value of e* - e can vary greatly throughout a season, Eq. [1] should be evaluated at short intervals, such as a day or a week, if it is used to predict growth. In this paper, e* - e is presented as the mean value for the daylight hours, following Bierhuizen and Slatyer (1965). Vapor pressure deficits were calculated for hourly intervals from the saturated vapor pressure at temperatures interpolated from the daily maximum and minimum and from the vapor pressure measured in a standard meteorological screen at the local recording time of 0900 h.
Water use efficiency refers to different processes and ratios in the literature. Here WUE refers to the ratio of yield to water used during crop growth, and TE refers to the ratio of yield to transpiration. We also refer to the crop management effects on WUE, meaning the practices under the control of farmers such as inputs, timing, tillage, and rotations. This review does not address genetic improvements to WUE, which are discussed by Richards et al. (2000) and Condon et al. (2000).
|
WATER USE EFFICIENCY AS A BENCHMARK
|
|---|
French and Schultz (1984) presented the results of agronomic experiments in South Australia by graphing grain yields in relation to the water use, which was defined as in Eq. [2] where the start of the period was sowing and the end was harvest. Figure 1 shows one of their graphs with the yields of experimental treatments plotted as points, and the water-limited potential yield plotted as a line; the intercept on the abscissa represents the Es during the growing season and the slope represents the maximum reported values of the TE (Hanks et al., 1969). The value of the slope was set at 20 kg ha-1 mm-1, based on the glasshouse results of Passioura (1976) and field results of Angus et al. (1980). The intercept on the abscissa, representing a seasonal Es of 110 mm, was estimated by placing the TE line along the frontier, or upper limit, of the experimental points. By using a single value for Es, French and Schultz (1984) overlooked the dependence of Es on the duration of incomplete cover and how frequently the soil is wet. Estimates of Es made in this way vary from 80 to 120 mm, typically representing 30% of the ET (Hanks et al., 1969; Angus et al., 1983; Hocking et al., 1997). Leuning et al. (1994) measured the Es during the late vegetative and reproductive stages of a wheat crop in southeastern Australia using microlysimeters and found that the Es represented 48% of the rainfall. The higher percentage from the direct measurement suggests that the intercept on the abscissa of the yield/ET graph may lead to an underestimate of the Es. French and Schultz (1984) made no allowance for the VPD in their analysis because there was relatively little variation between the seasons and localities, and crops in the region are grown over a similar time of year. In the example presented in Fig. 1, water use is represented only by growing-season rainfall because little water was stored during the fallow period in this environment.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 1. Relationship between wheat yield and water use from experiments conducted by French and Schultz (1984). The points represent yields measured in experiments conducted over many seasons and locations. The line represents the highest transpiration efficiency (TE) observed, and the intercept on the abscissa represents soil evaporation (Es). The vertical lines join data points representing different crop management treatments at a single experiment
|
|
While the TE line in Fig. 1 represents the water-limited yield (WLY), points under the line represent yields that are limited by genetic, environmental, or management factors other than water. The vertical lines joining points represent experimental treatments, such as different fertilizer applications, that lift yields towards the WLY. The vertical distance between a point and the TE line is the gap between the actual yield and the WLY. Failure to achieve the WLY leads to a wastage of water, for example, from additional Es.
The approach has had remarkable acceptance among wheat growers and advisers in the variable-rainfall environments of southeastern Australia as an indication of whether the crop yield is limited by the water supply or some other factor. The information helps them decide where improved management is needed.
The simplification of this method is that it uses a single estimate of the Es and neglects runoff, deep infiltration, and the timing of water supply relative to the demand of the crop. While runoff and deep infiltration are often negligible in the subhumid and semiarid environments of southeastern Australia, the assumption of an equal value of water supply at any time in the growing season is unjustified. Connor and Loomis (1991) showed that these simplifications can lead to significant errors in calculating the WLY.
The reason that Australian agronomists and wheat growers persist with the French–Schultz approach, despite the simplifications, is that errors in calculating the yield are minor compared with the yield gap between the WLY and actual yield (Angus et al., 1993). An example of the yield gap is shown in Fig. 2 where the mean annual wheat yields in the Wagga Wagga local government area are compared with the French–Schultz line and yields simulated with the SIMTAG (Simulation of Triticum Aestivum Genotypes) model (Stapper and Harris, 1989). This model simulates yield in relation to radiation, temperature, and daily water balance, and thus overcomes the deficiencies of the French–Schultz approach. However, over most of the range of observed yields, the SIMTAG and French–Schultz estimates are similar. At high yields, the SIMTAG estimates are mostly lower and more in accord with the observations of the maximum yields in this environment. The mean district yields are about 19% of the French–Schultz estimates and 36% of the SIMTAG estimates.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 2. Relationship between reported mean wheat yields in the Shire of Wagga Wagga for 1950 through 1992 in relation to growing-season rainfall and yields simulated by the SIMTAG model. The line represents the same transpiration efficiency (TE) and soil evaporation (Es) shown in Fig. 1
|
|
The conclusion from the analysis of French and Schultz is that water supply rarely limits the yield of commercial crops and that other environmental and management factors are normally limiting. This conclusion was initially surprising for the driest continent and has led to research on improving management methods rather than water limitations. The strength of the French–Schultz approach is that it can be readily calculated by farmers themselves without relying on more complex simulation models. For example, the WLY can be estimated from Eq. [3]:
 | (3) |
where 
, R is the growing-season rainfall (mm), and 
Es, the seasonal soil evaporation, equals 110 mm. Improvements have been made to estimate the water storage before sowing (Cornish and Murray, 1989) and with a computer-based version (D. Tennant, personal communication, 1994).
The gaps between actual and potential yields (Fig. 2) appear to be large compared with those in other regions. To some extent, the large gap may be due to inefficiencies associated with the extensive scale of wheat-growing on Australian farms. It is also due to the low inputs associated with the low ratio of grain price to cost per unit of input compared with the ratios in subsidized agricultural systems such as in North America and western Europe. Where the yield gaps are smaller and more difficult to identify, there may be advantages in using a more mechanistic approach to modeling, as suggested by Ritchie (1983).
The scatter of points in Fig. 1 and 2 fall below a frontier, or upper limit, and bear a striking resemblance to the graphs presented by Trumble (1937), which distinguish between yield limitations due to management and water supply, even though the values of yield and WUE are now higher. The similarity suggests there has been "re-search" in assessing water limitation for commercial crops, just as Tanner and Sinclair (1983) noted "re-search" for WUE. The French–Schultz approach also resembles the frontier and yield gap analysis used in economic studies (Kalirajan and Shand, 1994). This method estimates the magnitude and significance of the gap in yield between a particular crop and at an economic optimum. The unique feature of the method is that the economic frontier, or upper limit of efficiency, is based on a one-sided distribution curve, assuming that the upper limit has variation only on the lower side. This is similar in concept to the TE line in Fig. 1, but it has not been applied to the quantifying of agronomic limitations.
|
INCREASING WATER USE EFFICIENCY OR WATER USE?
|
|---|
Tanner and Sinclair (1983) concluded that a significant improvement in TE is unlikely. Ritchie (1983) considered that improving crop management, i.e., practices under the control of the farmer, can lead to an increased TE. Perhaps they have different perspectives because improving poor crop management may lead to the attainment of a maximum TE, represented by the line in Fig. 2, but further attempts at improving management have no effect. However, genetic improvements can increase the currently accepted maximum value of TE, based on selections for low 13C discrimination (Condon et al., 2000). There is also evidence that breeding for increased seedling vigor leads to more transpiration and an offsetting reduction in Es (Richards et al., 2000).
The other factor leading to increased TE is the increase in the atmospheric CO2 concentration (Polley, 2000). In the 17 yr since Tanner and Sinclair (1983) concluded that TE is a stable parameter, the level of CO2 in the atmosphere has risen from 335 to 360 ppm. For C3 species, TE is proportional to the CO2 gradient from the atmosphere to the mesophyll, which is about 70 ppm (Condon et al., 2000). The proportional change in the CO2 gradient between 1983 and 2000 is given by Eq. [4]:
 | (4) |
where Ca,2000 is the atmospheric CO2 concentration in 2000, Ca,1983 is the atmospheric CO2 concentration in 1983, and Ci is the CO2 concentration in the mesophyll. The proportional change in the CO2 gradient between 1983 and 2000 would lead to a 10% increase in the TE. This is supported by recent unpublished observations in southeastern Australia where TE is >22 kg ha-1 mm-1, rather than equal to 20 kg ha-1 mm-1 as reported by French and Schultz (1984).
Apart from these genetic and environmental effects, there is no evidence of crop management practices in southeastern Australia that lead to a greater TE. However, there is evidence of increased water use by crops in response to improved management using break crops and N fertilizer. Break crops such as canola (Brassica napus L.) and Indian mustard [Brassica juncea (L.) Czern. & Cosson] break the life cycle of cereal root diseases. Before the 1990s, most wheat in southeastern Australia was grown with little or no N fertilizer and in continuous cereal, fallow–cereal, or pasture–cereal rotations.
A summary of published results of field experiments shows that break crops and N fertilizer have affected water use and yield (Table 1). These were on-farm experiments in which the control treatment reflected commercial management at the time. Table 1 also presents the growing-season rainfall at each experimental site and the WUE based on the grain yield for the highest yielding treatment and the corresponding ET calculated from Eq. [2].
The apparent reason for the increased water use by wheat following break crops was a healthier root system following the control of cereal root diseases. The incidence and severity of pathogens of the root rot complex, principally take-all rot (Gauemannomyces graminis var. tritici) and crown rot (Fusarium graminearum), have been underestimated by most agronomists and farmers. Their importance has been shown by the large increases in wheat yield following Brassica spp. break crops (Angus et al., 1991; Kirkegaard et al., 1994). In 26 comparisons across southern Australia, wheat following canola outyielded wheat following wheat by an average of 19%, and the 2-yr gross margin of the canola–wheat sequence was 27% greater than the wheat–wheat sequence (Angus et al., 1999). A possible alternative explanation for the benefit of Brassica spp. break crops is biological drilling by the taproots, but Creswell and Kirkegaard (1995) could find no evidence for this. In addition to outyielding wheat following wheat, wheat following break crops in this environment also responds more to applied N (Angus et al., 1991).
In the studies reported in Table 1, crop water use was estimated from the rainfall plus the difference between the soil water at sowing and maturity using Eq. [2]. In all cases, the soil was sampled to 1.8 m with two cores per plot in each replicate of field experiments, dried at 105°C to constant weight, and expressed as volume of water using mean values of the bulk density obtained from the cores. Cores were preferred over neutron probes for measuring the soil water because the samples were also used to measure mineral N.
Eight of the 13 comparisons show significantly more water extraction by the crops that were better managed. Averaged over all of the comparisons, there was 23 mm more water used by the crops that were better managed. For the eight break crop comparisons, the mean additional water use was 31 mm, and for the five comparisons using additional N fertilizer, the mean additional water use was 10 mm. It is likely that a combination of both break crops and additional N would lead to even greater water use. The examples where the yield decreased in response to break crops and additional N represent haying off, which is discussed in the following section.
The profiles of soil water at one site show that additional water was extracted throughout the root zone, except in the top 10 to 20 cm (Fig. 3a). The matric potential of these samples was measured using the method of Greacen et al. (1987) in which filter papers are placed in sealed containers with soil and maintained at a constant temperature for 7 d. The change in weight was then used to calculate the matric potential based on a calibration using a dew point hygrometer (A.F. van Herwaarden and J.S. Boyer, unpublished data, 1995). The soil matric potential at the time of sowing decreased with depth from -0.02 to -0.12 MPa at the time of sowing. By maturity, both high- and low-N crops dried the soil to matric potentials <-3 MPa where no fertilizer N had been applied and to -5 MPa where 80 kg N ha-1 had been applied (Fig. 3b). Brown (1971) also reported additional water extraction in response to applied N, but in his case, wheat that was supplied with N fertilizer extracted water from deeper soil layers.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 3. Soil profiles of (a) water content and (b) matric potential measured at sowing and maturity of wheat growing at Pucawan, Australia at two levels of applied N. The matric potential data are as presented by van Herwaarden et al. (1998a), and the previously unpublished soil water data refer to the same samples. The total water content in each profile is also shown in (a) along with the least significant differences between them
|
|
These results impinge on the long-running controversy about the generality of a permanent wilting point, measured with a pressure plate, above which plants could not maintain normal growth. Results reviewed by Veihmeyer and Hendrickson (1950) showed that crops did not extract soil water below a matric potential of -1.5 MPa while Slatyer (1957) presented evidence of lower and more variable values. Kramer (1969) suggested that even if the permanent wilting point was <-1.5 MPa matric potential, it was still an important soil value because there was little change in the soil water content with relatively large differences in the matric potential, owing to the shape of the curve relating the matric potential to the water content.
In this study, values of the soil matric potential between -3 and -5 MPa for soil depths 30 to 100 cm show that a permanent wilting point of -1.5 MPa is by no means general (Fig. 3b). For this soil, the additional volume of water extracted to matric potentials between -3 and -5 MPa (Fig. 3a) is sufficient to contribute significantly to yield. The soil type at Pucawan was a red-brown earth, one of the most widely distributed cropping soils in southeastern Australia, classified as Dr2.23 (Northcote, 1979) or Rhodoxeralf (Soil Survey Staff, 1975). The layers between 30 and 100 cm contain about 50% clay, so it is feasible that the large difference in the matric potential is sufficient to account for the difference in the 1 to 2% volume extracted by the more vigorous crops.
Ritchie (1981) proposed an alternative to estimating the permanent wilting point with a pressure plate. His approach was to measure the lower limit of soil water content in the field after a crop matured during a terminal drought. It overcame the unrealistic assumption of the earlier method that roots can dry the subsoil as much as the topsoil even though they are present for a shorter time. Ritchie (1981) reviewed examples where nutrient deficiency affected the lower limit, but he implied that these were exceptions to a general lower limit for a soil. The results presented here do not support the suggestion that the lower limit is a unique characteristic of a soil or even of a soil–species combination, but they suggest that it is a variable influenced by the vigor of the growing crop.
While there is good evidence of additional soil water extraction by better management of a single crop, it is not yet clear whether that water accumulates over one or more years, depending on the rainfall, and is not replenished after better management is introduced. One possibility is that there is increased water use by one crop in the transition to improved crop management. Another is that a drier soil will lead to deeper infiltration of rain, and hence less loss due to runoff and evaporation. In this case, greater crop water use could be expected to continue as long as a high level of crop management is maintained.
An implication of a crop extracting more soil water is that, in some circumstances, there will be less residual soil water and an increased risk of water deficit for subsequent crops. Another is that a drier soil profile will retain more rainfall than a wetter profile, and thus reduce the risk of groundwater recharge. Where the groundwater is salty, the use of more water by vigorous crops reduces the risk of salinity.
|
HAYING OFF AND WATER USE EFFICIENCY
|
|---|
In southeastern Australia, dryland wheat is sown in late autumn or early winter and completes its vegetative stage during mild winter conditions when a water deficit is unusual. The stage from stem elongation to anthesis occurs during early spring when the water supply varies greatly from year to year. The grain-filling period occurs during late spring when the evaporative demand normally exceeds rainfall.
Management to increase crop growth in the vegetative phase during mild winter weather might be expected to increase yield because of higher WUE when the evaporation and VPD are low and because Es would be reduced by additional canopy cover. Daily pan evaporation for Temora in 1991 varied between 0.0 and 7.2 mm before anthesis and between 1.4 and 13.6 mm after anthesis. The mean daily VPD varied between 0.1 and 1.9 kPa before anthesis and between 0.5 and 3.2 kPa after anthesis (Fig. 4).
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 4. (a) Class A pan evaporation and (b) vapor pressure deficit (VPD) for daylight during the growing season of wheat at Temora, Australia in 1991
|
|
A possible limitation of this hypothesis is that additional vegetative growth can lead to haying off, a yield reduction in response to additional N, when vigorous vegetative growth is followed by a terminal drought. Haying off is relatively common in southeastern Australia, but it has also been reported in other dryland cropping regions (van Herwaarden et al., 1998a). The common explanation for the yield reduction has been a lack of soil water during grain filling because more soil water was used in producing additional vegetative material. Research by van Herwaarden et al. (1998b) suggests that a more important reason for the yield reduction is a shortage of soluble carbohydrate available for retranslocation. This shortage comes about because of a negative association between N and soluble carbohydrate. This relationship is commonly observed in growing crops and is not necessarily related to a water deficit (e.g., Batten et al., 1993). There are two examples of haying off in the data shown in Table 1. At Junee, the yields of wheat growing after break crops were lower than those after wheat and at Wagga Wagga, the yield of wheat supplied with 200 kg N ha-1 was lower than that of the crop with no added N. This N input is unrepresentative of the normal practice in the region and is reported here because the soil water content was measured for this treatment but not for treatments supplied with normal N applications. The original report shows that all N inputs, starting with 40 kg N ha-1, led to progressive yield reductions in this experiment (van Herwaarden et al., 1998a).
The implications for WUE from these insights into haying off and retranslocation are explored using budgets of water and dry matter in relation to the N status of wheat crops grown at Pucawan (van Herwaarden et al., 1998a,b). The budgets in Table 2 show the partition of ET into transpiration and Es using the method of Cooper et al. (1983) for the phases before and after anthesis. Briefly, this method uses fitted curves of ET (from the water balance), intercepted radiation (from regular measurements), and surface water status to provide daily interpolated values of Es. Transpiration is then calculated as the difference between ET and Es. The dry matter data used in calculating TE were obtained from van Herwaarden et al. (1998a). The VPD values represent the means of the data shown in Fig. 4b; the values after anthesis are triple those before anthesis.
View this table:
[in this window]
[in a new window]
|
Table 2. Partition of water use and calculation of water use efficiency (WUE) before and after anthesis in relation to vapor pressure deficit (VPD) for wheat growing at Pucawan with low (no applied N) and high (80 kg N ha-1) N status
|
|
As expected, the values of TE were greater before anthesis than after anthesis because of the lower VPD. However, the k value was not constant between the phases as would be expected from Eq. [1], and the transpiration after anthesis was more efficient in producing yield than would be expected from a constant value of k. The differences in k may reflect errors in partitioning transpiration and Es or the methods lack of accounting for growth responses to radiation and temperature. Surprisingly, the values of TE are lower for the high-N status crops. This result differs from those reviewed by Ritchie (1983), which showed an increased TE at high-N status, and may reflect consequences of haying off. However, a reduction in the TE of wheat due to a high-N status that is not associated with haying off has also been observed on lysimeters at Wagga Wagga (F.X. Dunin, personal communication, 2000).
The value of the increased ET before anthesis for grain yield depends on the retranslocation of previously assimilated carbohydrate and protein to grain while increased ET after anthesis leads directly to additional yield. The relative value of an additional 1 mm of ET before and after anthesis is explored in Table 3 based on the data in Table 2. Using the TE and Es values from Table 2, the additional biomass produced for an additional 1 mm of ET is allocated in Table 3 to water soluble carbohydrate, protein, and (by difference) structural biomass at anthesis, based on the proportions of each reported by van Herwaarden et al. (1998b). Assuming that water soluble carbohydrate and protein are available for retranslocation, each 1 mm of ET before anthesis led to an additional 6 to 7 kg ha-1 of grain growth. In contrast, the conservation of water for ET after anthesis produced dry matter with an efficiency of 35 to 37 kg ha-1 mm-1 and grain with an efficiency of 33 kg ha-1 mm-1. The difference between the efficiencies for the total dry matter and grain was due to an accumulation of cellulose in the stems during grain filling.
View this table:
[in this window]
[in a new window]
|
Table 3. Comparison of the effects of 1 mm of evapotranspiration (ET) before or after anthesis on biomass production of low (no applied N) and high (80 kg N ha-1) wheat crops, as described in Table 2
|
|
These results show that increasing ET during the vegetative phase was a less effective way to increase yield than delaying the same increase in ET until grain filling when assimilates were mostly allocated to grain. The advantage of a higher TE during the vegetative phase, associated with a lower VPD, was offset by the allocation of the growth to the structural tissue, which was unavailable for retranslocation to grain. The lower TE during grain filling was more than compensated for by the larger proportion of transpiration to ET and efficient allocation of additional growth to grain. In other words, there was a disadvantage in modifying crop management so that more water was transpired before anthesis, and the disadvantage was more pronounced when the additional transpiration was due to a higher N status. The crops at Pucawan did not hay off badly, as shown in Table 2 by the small effect of the N status on the soil water available for transpiration after anthesis. In situations where a crop uses more soil water before anthesis because of a high-N status, the disadvantage of a high-N status would be greater.
|
CONCLUSIONS
|
|---|
There is no evidence from the results presented here that improved crop management can increase TE beyond a maximum such as that expressed by the lines in Fig. 1 and 2. Previous studies reviewed by Ritchie (1983) showed that applied N and P lead to increased TE, presumably by increasing the value towards the maximum. However, the results presented here show that improving crop management with break crops and N fertilizer leads to more soil water extraction by a crop maturing during a terminal drought. The WLY defined by the maximum TE and Es is a valuable benchmark for grain growers in dryland regions to quantify management limitations to yield.
Three factors that influence the relative TE before and after anthesis are the proportion of transpiration to Es, VPD, and the proportion of soluble carbohydrate that is allocated to grain. In southeastern Australia, additional preanthesis growth leads to additional transpiration relative to Es and to a higher TE associated with a lower VPD. However, these advantages may be more than offset because the additional growth is directed to structural material rather than soluble carbohydrate that can be retranslocated to grain after anthesis. Increasing the vegetative growth with a high-N status leads to a strong allocation of assimilate to structural tissue, and therefore to a potential shortage of soluble carbohydrate for retranslocation to grain. The alternative of conserving water for ET after anthesis is a more promising means of increasing yield, despite a lower TE, because transpiration leads directly to an increase grain growth, as Passioura (1976) showed. It is also consistent with Fischer (1979), who concluded that excessive dry matter at anthesis was undesirable in this environment.
The value of increased soil water extraction and delayed water use until after anthesis will depend on seasonal conditions. The long-term benefits for the defined environment could be evaluated with a simulation model that included suitable functions for the soil water extraction in relation to the crop management and retranslocation of assimilates.
|
ACKNOWLEDGMENTS
|
|---|
We are grateful to the Grains Research and Development Corporation for funding support and to Tony Condon for helpful comments. Figure 1 is reproduced with permission from the Australian Journal of Agricultural Research, and Fig. 2 is reproduced with permission from the Journal of Agricultural Meteorology. We acknowledge collaborating farmers for providing land for the field experiments.
Received for publication February 14, 2000.
|
REFERENCES
|
|---|
- Angus, J.F., J.M. Desmarchelier, P.A. Gardner, A. Green, P.J. Hocking, G.N. Howe, J.A. Kirkegaard, S. Marcroft, J.A. Mead, G.D. Pitson, T.D. Potter, M.M. Ryan, M. Sarwar, A.F. van Herwaarden, and P.T.W. Wong. 1999. Canola and Indian mustard as break crops for wheat [CD-ROM computer file]. Proc. Int. Rapeseed Congr., 10th, Canberra, ACT, Australia. 26–29 Sep. 1999. Groupe Consultatif Int. de Recherche sur le Colza, Paris.
- Angus, J.F., M. Stapper, and J.R. Donnelly. 1993. Simulation models for strategic and tactical management of crops and pastures. J. Agric. Meteorol. 48:775–778.
- Angus, J.F., A.F. van Herwaarden, and G.N. Howe. 1991. Productivity and break crop effects of winter-growing oilseeds. Aust. J. Exp. Agric. 31:669–677.
- Angus, J.F., H.A. Nix, J.S. Russell, and J.E. Kruizinga. 1980. Water use, growth, and yield of wheat in a subtropical environment. Aust. J. Agric. Res. 31:873–886.
- Angus, J.F., S. Hasegawa, T.C. Hsaio, S.P. Liboon, and H.G. Zandstra. 1983. The water balance of post-monsoonal dryland crops. J. Agric. Sci. (Cambridge) 101:699–710.
- Batten, G.D., A.B. Blakeney, V.B. McGrath, and S. Ciavarella. 1993. Non-structural carbohydrate: Analysis by near infrared reflectance spectroscopy and its importance as an indicator of plant growth. Plant Soil 155/156:243–246.
- Bierhuizen, J.F., and R.O. Slatyer. 1965. Effect of atmospheric concentration of water vapour and CO2 in determining transpiration–photosynthesis relationships of cotton leaves. Agric. Meteorol. 2:259–270.
- Briggs, L.J., and H.L. Shantz. 1916. Daily transpiration during the normal growth period and its correlation with the weather. J. Agric. Res. (Washington, DC) 7:155–212.
- Brown, P.L. 1971. Water use and soil water depletion by dryland winter wheat as affected by nitrogen fertilization. Agron. J. 63: 43–46.
- Condon, A.G., R.A. Richards, G.J. Rebetzke, and G.D. Farquhar. 2000. Improving intrinsic water-use efficiency and crop yield. Crop Sci. (in press).
- Connor, D.J., and R.S. Loomis. 1991. Strategies and tactics for water-limited agriculture in low-rainfall mediterranean climates. p. 441–465. In Proc. Int. Symp. on improvement and management of winter cereals under temperature, drought, and salinity stresses, Cordoba, Spain. 26–29 Oct. 1987. Instituto Nacional de Investigaciones Agrarias, Madrid, Spain.
- Cooper, P.J.M., J.D.H. Keatinge, and G. Hughes. 1983. Crop evapotranspiration—a technique for calculation of its components by field measurements. Field Crops Res. 7:299–312.
- Cornish, P.S., and G.M. Murray. 1989. Low rainfall rarely limits wheat yields in southern New South Wales. Aust. J. Exp. Agric. 29:77–83.
- Cresswell, H.P., and J.A. Kirkegaard. 1995. Subsoil amelioration by plant roots—the process and the evidence. Aust. J. Soil Res. 33: 221–239.
- de Wit, C.T. 1958. Transpiration and crop yields. Versl. Landbouwk. Onderz. 64.6. Inst. of Biol. and Chem. Res. on Field Crops and Herbage, Wageningen, The Netherlands.
- Fischer, R.A. 1979. Growth and water limitation to dryland wheat yield in Australia: A physiological framework. J. Aust. Inst. Agric. Sci. 45:83–94.
- French, R.J., and J.E. Schultz. 1984. Water use efficiency in wheat in a mediterranean type environment: II. Some limitations to efficiency. Aust. J. Agric. Res. 35:765–775.
- Greacen, E.L., G.R. Walker, and P.G. Cook. 1987. Evaluation of the filter paper method for measuring soil water suction. p. 137–143. In Proc. Int. Conf. Measurement of Soil and Plant Water Status, Logan, UT. 6–10 July 1987. Utah State Univ., Logan.
- Hanks, R.J., H.R. Gardner, and R.L. Florian. 1969. Plant growth–evapotranspiration relations for several crops in the central great plains. Agron. J. 61:30–34.[Abstract/Free Full Text]
- Hocking, P.J., J.A. Kirkegaard, J.F. Angus, A.H. Gibson, and E.A. Koetz. 1997. Comparison of canola, Indian mustard, and Linola in two contrasting environments: I. Effects of nitrogen fertilizer on dry-matter production, seed yield and seed quality. Field Crops Res. 49:107–125.
- Kalirajan, K.P., and R.T. Shand. 1994. Economics in Disequilibrium. Macmillan India, New Delhi, India.
- Kirkegaard, J.A., P.A. Gardner, J.F. Angus, and E. Koetz. 1994. Effect of Brassica break crops on the growth and yield of wheat. Aust. J. Agric. Res. 45:529–545.
- Kramer, P.J. 1969. Plant and soil relationships: A modern synthesis. McGraw-Hill, New York.
- Leuning, R., A.G. Condon, F.X. Dunin, S. Zegelin, and O.T. Denmead. 1994. Rainfall interception and evaporation from soil below a wheat canopy. Agric. For. Meteorol. 67:221–238.
- Northcote, K.H. 1979. A factual key for the recognition of Australian soils. Rellim Tech. Publ., Adelaide, SA, Australia.
- Passioura, J.B. 1976. Physiology of grain yield in wheat growing on stored water. Aust. J. Plant Physiol. 3:559–565.
- Polley, H.W. 2000. Implications of climate change for crop yield and water-use efficiency. Crop Sci. (in press).
- Richards, R.A., G.J. Rebetzke, A.G. Condon, and A.F. van Herwaarden. 2000. Breeding opportunities for increasing efficiency of water use and crop yield in temperate cereals. Crop Sci. (in press).
- Richardson, A.E.V. 1923. The water requirements of farm crops. J. Dep. Agric. Victoria Aust. 21:193–212, 257–284, 321–339, 385–404, 449–477.
- Ritchie, J.T. 1981. Soil water availability. Plant Soil 58:327–338.
- Ritchie, J.T. 1983. Efficient water use in crop production: Discussion on the generality of relations between biomass production and evapotranspiration. p. 29–44. In H.M. Taylor et al. (ed.) Limitations to efficient water use in crop production. ASA, Madison, WI.
- Slatyer, R.O. 1957. The significance of the permanent wilting percentage in studies of plant and soil water relations. Bot. Rev. 23:585–636.
- Soil Survey Staff. 1975. Soil Taxonomy. USDA Agric. Handb. 436.
- Stapper, M., and H.C. Harris. 1989. Assessing the productivity of wheat genotypes in a mediterranean climate, using a simulation model. Field Crops Res. 20:129–152.
- Tanner, C.B., and T.R. Sinclair. 1983. Efficient water use in crop production: Research or re-search. p. 1–27. In H.M. Taylor et al. (ed.) Limitations to efficient water use in crop production. ASA, Madison, WI.
- Taylor, H.M., W.A. Jordan, and T.R. Sinclair. 1983. Limitations to efficient water use in crop production. ASA, Madison, WI.
- Trumble, H.C. 1937. The climatic control of agriculture in South Australia. Trans. R. Soc. South Aust. 61:41–62.
- van Herwaarden, A.F., J.F. Angus, G.D. Farquhar, G.N. Howe, and R.A. Richards. 1998a. Haying-off, the negative grain yield response of dryland wheat to nitrogen fertilizer: I. Biomass, grain yield, and water use. Aust. J. Agric. Res. 49:1067–1081.[ISI]
- van Herwaarden, A.F., J.F. Angus, R.A. Richards, and G.D. Farquhar. 1998b. Haying-off, the negative grain yield response of dryland wheat to nitrogen fertilizer: II. Carbohydrate and protein dynamics. Aust. J. Agric. Res. 49:1083–1093.[ISI]
- van Herwaarden, A.F., G.D. Farquhar, J.F. Angus, and R.A. Richards. 1996. Physiological responses to six spring wheat varieties to nitrogen fertilizer. p. 570–573. In M. Asghar (ed.) Proc. Australian Agron. Conf., 8th, Toowoomba, QLD, Australia. 30 Jan.–2 Feb. 1996. Australian Soc. of Agron., Toowoomba.
- Veihmeyer, F.J., and A.H. Hendrickson. 1950. Soil moisture in relation to plant growth. Annu. Rev. Plant Physiol. 1:285–304.
This article has been cited by other articles:
|
|
|
|
 
D. J. Lyon, D. C. Nielsen, D. G. Felter, and P. A. Burgener
Choice of Summer Fallow Replacement Crops Impacts Subsequent Winter Wheat
Agron. J.,
March 12, 2007;
99(2):
578 - 584.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. J. Lopez-Bellido, L. Lopez-Bellido, J. Benitez-Vega, and F. J. Lopez-Bellido
Tillage System, Preceding Crop, and Nitrogen Fertilizer in Wheat Crop: II. Water Utilization
Agron. J.,
January 1, 2007;
99(1):
66 - 72.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. M. Impa, S. Nadaradjan, P. Boominathan, G. Shashidhar, H. Bindumadhava, and M. S. Sheshshayee
Carbon Isotope Discrimination Accurately Reflects Variability in WUE Measured at a Whole Plant Level in Rice
Crop Sci.,
October 27, 2005;
45(6):
2517 - 2522.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. J. Schlegel, C. A. Grant, and J. L. Havlin
Challenging Approaches to Nitrogen Fertilizer Recommendations in Continuous Cropping Systems in the Great Plains
Agron. J.,
March 1, 2005;
97(2):
391 - 398.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. L. Anderson
Are Some Crops Synergistic to Following Crops?
Agron. J.,
January 1, 2005;
97(1):
7 - 10.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. E. Abbate, J. L. Dardanelli, M. G. Cantarero, M. Maturano, R. J. M. Melchiori, and E. E. Suero
Climatic and Water Availability Effects on Water-Use Efficiency in Wheat
Crop Sci.,
March 1, 2004;
44(2):
474 - 483.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
V. O. Sadras and D. K. Roget
Production and Environmental Aspects of Cropping Intensification in a Semiarid Environment of Southeastern Australia
Agron. J.,
January 1, 2004;
96(1):
236 - 246.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. A. Richards, G. J. Rebetzke, A. G. Condon, and A. F. van Herwaarden
Breeding Opportunities for Increasing the Efficiency of Water Use and Crop Yield in Temperate Cereals
Crop Sci.,
January 1, 2002;
42(1):
111 - 121.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. G. Condon, R. A. Richards, G. J. Rebetzke, and G. D. Farquhar
Improving Intrinsic Water-Use Efficiency and Crop Yield
Crop Sci.,
January 1, 2002;
42(1):
122 - 131.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
