Previous PageTable Of ContentsNext Page

Developing testable hypotheses on the impacts of sub-soil constraints on crops and croplands using the cropping systems simulator APSIM.

Zvi Hochman1, Merv Probert2 and Neal P. Dalgliesh1

1 CSIRO Sustainable Ecosystems/APSRU, PO Box 102 Toowoomba, Qld 4250. www.csiro.au.
Email Zvi.Hochman@csiro.au; Neal.Dalgliesh@csiro.au
2
CSIRO Sustainable Ecosystems/APSRU, Queensland Bioscience Precinct, 306 Carmody road, St Lucia, Qld 4067.
Email Merv.Probert@csiro.au

Abstract

A simulation model was used to develop hypotheses about the impacts of sodicity and salinity on wheat crops, runoff and drainage in Australia’s northern cropping region. Based on the hypothesis that subsoil constraints reduce crop yield by restricting their roots’ access to soil moisture and nutrients, a range of levels of subsoil constraints was simulated by progressively restricting plant available water capacity of a typical Grey Vertosol at four locations representing the range of rainfall distributions in the region. While predicted impacts of subsoil constraints were highly variable from year to year, 100 year means showed that these impacts on yield, drainage and runoff were greater when pre-season soil moisture was high. Impacts were similar at all sites while their absolute values were proportional to average annual rainfall and summer dominance.

Media summary

Simulation was used to investigate the impacts of sodicity and salinity in Australia’s northern grains region. Large impacts on grain yield, drainage and runoff are predicted.

Key Words

Salinity, sodicity, simulation, crop, drainage, runoff

Introduction

It is estimated that over 0.4Mha in the Australian Northern Cropping region are affected by a combination of high levels of subsoil sodicity and salinity. These soils are observed to have poor crop yields associated with reduced rooting depth and stored soil moisture (So and Aylmore 1993). The problem of subsoil constraints (SSCs) is the current focus of a significant research effort involving several research agencies in NSW and Qld. The aim of this study is to use a cropping system simulator (APSIM; Keating et al. 2003) to develop testable hypotheses that would contribute to the design of field experiments.

The guiding hypothesis

Let us start with the proposition that SSCs associated with high levels of salinity and sodicity reduce crop yields primarily by restricting a crop’s access to soil moisture and nutrients. This restriction may be achieved by a number of mechanisms such as restricting the rate of root front development, impairing of root function, or increasing osmotic pressure. Water drainage in the subsoil may also be affected by low hydraulic conductivity. The first hypothesis we propose is that the net impact of these mechanisms can be accounted for by consideration of their effect on the lower limit of a crop’s available soil water (crop lower limit or CLL) which reduces the plant available water capacity (PAWC) of a soil. The value of this hypothesis is that CLL can be readily estimated in the field (Dalgliesh and Foale 1998) and that it can be readily tested by gathering a minimum data set for crop simulation on a sufficient number of sites representing a range of severities of SSCs.

Grey Vertosols are the most important soil type on which SSC occur in the northern cropping zone and therefore wheat cropping on such soils serves as the model system for this paper. In Grey Vertosols we typically observe gradual increases of salinity and sodicity with depth of soil. We also observe a gradual increase in CLL with depth. We can consider the severity of subsoil constraints as departures from the CLL that may be observed on a typical Grey Vertosol (i.e. without SSCs). Analysis of a large data set of such soils in the Northern Grain Zone has provided a soil moisture characterisation of such a soil (Hochman et al. 2001). Figure 1 illustrates the presumed departure with depth of 3 nominated levels of SSCs relative to the drained upper limit (DUL) and CLL of a typical grey vertosol.

Figure 1. Assumed impact of 3 levels of SSCs on wheat CLL of a Grey Vertosol.

Predicted impacts of SSC on crop yields

Given that experimental results will be obtained in a limited sample of specific seasons, it seems desirable to use a cropping systems simulator to investigate the seasonal variability of the impact of SSCs on grain yield. Season variability could be considered to have two sources: 1. difference in amount of stored soil moisture that can be exploited by the crop, and 2. difference in effective use of in season rainfall. To investigate in-season differences we can reset stored soil moisture at the start of each season (e.g. 30th April) and take the past 100 years’ weather data for a given location and compare the simulated grain yields of wheat that are calculated with and without the limitations of SSCs on CLL. Figure 2 illustrates the wheat grain yield differences between a crop growing in Goondiwindi on an unconstrained Grey Vertosol without any SSCs (SSC 0) and the same crop growing on a soil that has high SSCs (SSC 3) as described in Figure 1 above. In both cases pre season soil moisture is reset at the start of each season to be 2/3 full relative to the unconstrained Grey Vertosol. As yield differences between treatments vary from less than 200 kg/ha in some years to nearly 3 t/ha in others, we can expect yield responses to be highly variable from year to year.

To investigate the impact of stored moisture we contrast the mean results of the 4 different levels of SSCs represented in figure 1 at two pre season soil moisture profiles, 1/3 full versus 2/3 full. While both scenarios show progressive yield losses in response to severity of SSCs, the effect on wheat grain yield is greater where pre-growing season stored moisture, and consequently yield potential, is greater (Figure 3). We conclude by postulating that both pre-season (Figure 3) and in season conditions (Figure 2) modify the crop’s response to SSC.

Impacts of SSCs on the soil resource – runoff and drainage

In addition to the impact on crop yield we are also interested on the impact of SSCs on the soil resource. To this purpose we can use the simulator to investigate the impacts of SSCs on runoff and drainage. We investigated the impacts of SSCs on drainage and runoff as influenced by pre-growing season soil moisture. For drainage, subsoil constraints had an important impact on average annual drainage in the high starting soil moisture scenario, but much less effect when starting soil moisture was only 1/3 PAWC (Figure 4). A similar interaction between SSCs and pre season soil moisture was observed for runoff (Figure 5). Interestingly, less runoff is predicted with higher pre-season starting moisture, a counterintuitive result which arises from the amount of residues remaining from the previous crop affording protection from runoff over the fallow period.

Figure 2. Differences in annual wheat yields between a healthy Grey Vertosol (SSC 0) and one subject to severe SSCs at Goondiwindi. Each season starts with a 2/3 full moisture profile.

Figure 3. Average wheat yield responses to increasing SSCs at contrasting pre-season soil moisture levels at Goondiwindi.

Figure 4. Drainage response to SSCs and pre-season soil moisture.

Figure 5. Runoff response to SSCs and pre-season soil moisture.

Impacts of SSCs at different locations

Also of interest is the extent to which these impacts are modified by location. Here we compare four locations: Gunnedah, Goodiwindi, Roma, and Emerald with similar average annual rainfalls (608mm, 581mm, 585mm and 624mm respectively) but contrasting in degrees of dominance of summer rainfall distributions. Figure 6 shows that sites with more summer-dominant rainfall had lower yields, although the response to SSCs is similar at all sites. As with yield, the impact of SSCs on drainage is similar for all sites (Figure 7). Both average annual rainfall (aar) and summer dominance influence the overall drainage trend: drainage at Emerald > Goondiwindi > Roma is consistent with higher aar leading to higher drainage at these sites. However, Gunnedah with the lowest aar has a similar drainage trend as Goodiwindi. Here the aar trend is moderated by a trend for lower overall drainage in the more summer-dominant rainfall. This trend may be explained by 1. Rainfall is more concentrated in the high evaporation potential season, and 2. Summer rainfall events tend to be more concentrated than winter rainfall (thus excess water is more likely to be expressed in runoff). Site differences in runoff (not shown) are consistent with summer dominance (and tendency for more concentrated rainfall events). Response to reduced PAWC is relatively flat indicating that runoff is less responsive to subsoil conditions.

Figure 6 Wheat yield response to SSCs and Location (summer rain dominance) Pre season
Soil Moisture = 2/3 Full

Figure 7. Drainage Response to SSCs and Location (summer rain dominance). Pre season Soil Moisture = 2/3 Full.

Conclusion

This simulation study is based on the assumption that the impacts of SSCs on crop growth and soil water processes can be accounted for by their impacts on PAWC. This assumption is yet to be tested, but it provides a useful working hypothesis and a starting point for further investigation. Postulated impacts of SSCs are:

  • The impact of subsoil constraints on yield potential is highly variable from season to season
  • The impacts of subsoil constraints on yield potential can be expected to be greater where pre-season soil moisture is high
  • SSCs can be expected to increase both runoff and drainage beyond the reach of crop roots, especially when pre-season soil moisture is higher
  • Similar wheat yield responses to reduced PAWC may be expected over a range of sites representing the range of summer-winter rainfall distributions in the Northern Grain Zone.
  • Under a wheat crop, similar drainage responses to reduced PAWC may be expected over a range of sites representing the range of summer-winter rainfall distributions in the North.
  • Under a wheat crop, runoff responses were proportional to summer dominance. Similar, relatively flat, runoff responses to reduced PAWC were observed for all sites with the least summer-dominant site of Gunnedah showing the flattest response to reduced PAWC.

Acknowledgement

This work was funded by the GRDC strategic initiative on Combating Sub-Soil Constraints.

References

Dalgliesh NP and Foale MA (1998). Soil Matters – monitoring soil water and nitrogen in dryland farming. Agricultural Production Systems Research Unit, Toowoomba, Australia. 122 pp.

Hochman Z Dalgliesh NP and Bell K (2001). Contributions of soil and crop factors to plant available soil water capacity of annual crops on black and grey vertosols. Australian Journal ofAgricultural Research 52, 955-961

Keating BA Carberry PS Hammer GL Probert ME Robertson MJ Holzworth D Huth NI Hargreaves JNG Meinke H Hochman Z McLean G Verburg K Snow V Dimes JP Silburn M Wang E Brown S Bristow KL Asseng S Chapman S McCown RL Freebairn DM and Smith CJ (2003). An overview of APSIM, a model designed for farming systems simulation. European Journal of Agronomy 18, 267-288

So HB and Aylmore LAB (1993). How do sodic soils behave? The effects of sodicity on soil physical behaviour. Australian Journal of Soil Research 31, 761-777.

Previous PageTop Of PageNext Page