Effect of row spacing, deep ripping, gypsum, water harvesting furrow and mulching crops grown on hard setting sodic soils

Key Messages

• Depth of sowing is critical and difficult to control on deep ripped/cultivated heavy-textured soils.
• Crop establishment was negatively affected by the deep ripping treatments, mainly due to cloddy seedbeds and furrow infill.
• Wide row spacing (60 cm) has no yield benefits over narrow row spacing (30 cm), and treatments effects depend on seasonal variability.

Aims

Heavy-textured soils in the eastern grainbelt are subject to a range of constraints including sodicity, compaction and transient salinity. Each of these reduces plant growth by restricting water infiltration, water storage, crop emergence and root growth. Petersen (2016) estimates that sodicity alone results in $580 million in lost production in WA. Strategies to improve water storage and availability to plants include combinations of deep tillage, gypsum, wide row spacing, furrow forming wheels and stubble mulching. This paper reports on a four-year trial aimed at developing and assessing improved management systems to increase water storage in hard-setting sodic soils.

Method

The field trial site was located at Merredin Research Station (MRS) in paddock 3c (S31.505, E118.2126). The soil at the trial site is classified as a calcareous loamy earth (Schocknecht and Pathan, 2013) with multiple constraints, which include hardsetting, subsoil compaction and sodicity. There was considerable variation in physical and chemical properties at different depths (Table 1). Top soil was comprised primarily of sandy clay particles (clay 34-44%) and subsoil medium clay (clay 46-52%). A hard setting surface, poorly drained and few lime fragments (pedogenic carbonates) have been observed at >30 cm. Subsoils are highly compacted with soil resistance more than 3.5MPa within 20-25cm.The topsoil (0-20 cm) is moderately neutral, non-saline, with low soil organic carbon. The subsoil (20-80 cm) is alkaline, moderately saline (EC>0.45 mS/m) and highly sodic (ESP>15%) at depth (Table 1).

Table 1: Selected soil chemical properties of the trial site (MRS paddock 3c) at different depths and samples were taken prior to applying treatments.

The deep ripping (~40 cm) treatment (four 26 tines spaced 445 mm apart with five leading tines100 mm shallower and 223 mm off line) was completed in April 2015. Gypsum (2.5 t/ha; 84% purity) was applied on the surface by hand prior to deep ripping in 2015 and reapplied in April 2018. Furrow (5.0 cm height and 23.1-degree angle) was made using especially deigned wheels (23.5 cm wide and 5.0 cm deep) attached behind the tractor for harvesting water in the row. Following the 2015 growing season, each field plot (45 m) was split with half the plot across the entire area receiving a mulch of wheat straw (5 t/ha) and half without mulch. The narrow (30 cm) and wide row (60 cm) spacing can be combined with the five major treatments, and mulching effect can be evaluated within these treatments (Figure 1). Ten treatments were laid out in a factorial randomised block design with three replicates.

Crops were sown using knifepoints, press wheels and standard agronomic practices for the region. The wheat straw mulch was applied using hand immediately after seeding. The mineral nutrient analysis of the wheat straw mulch used in this trial are given below (Table 2). Necessary fertilise (such as K-Till Extra, Agras, AgStarter 50-60kg/ha) and herbicide was applied as standard regional practice for specific crops. Measurements made during the growing season included germination counts, crop growth/vigour (NDVI), shoot dry matter, grain yield and quality (Table 3).

Table 2: Mineral nutrient analysis of the wheat straw mulch used in the trial.

Table 3: A summary of the trial site major activities during last four years (2015-2018).

Results

Crop establishment (plants/m²) was negatively affected by the deep ripping treatments resulting in reduced plant numbers, irrespective of row spacing and mulching for the years 2015 to 2017 (Table 4). For example, in 2016 growing season, 61-76 plants/m² were established in the un-ripped treatment, compared with 44-61 plants/m² in the deep ripped treated plots, irrespective of mulching. This is mainly due to the seed being sown too deep (>5 cm) in the soft soils and exacerbated by large clods and furrow infill. Deep ripping can also reduce soil water content in the surface layers by disturbing and exposing the soil to evaporation, which can reduce crop establishment and grain yield in drier season. Depth of sowing is critical and difficult to control on ameliorated/cultivated soils were also observed other trials in south Western Australia (Blackwell et al, 2016).

Table 4: Effect of treatments on crop establishment (plant/m²) during last four years (2015-2018), irrespective of mulching.

Despite poorer crop establishment, plant shoot dry matter and vigour (NDVI) were not significantly affected by the deep ripping treatments, irrespective of row spacing and mulching. Significant growth and shoot dry matter was observed in the year of lowest rainfall (319-322 mm) during 2017 and 2018 growing season under mulch comparer to no mulch regardless treatments (data not shown).

Table 5: Effect of various treatments on grain yields (t/ha) during last four years (201-2018), irrespective of mulching.

Grain yield responses to deep ripping were temporally and seasonally variable (Table 5). Deep ripping reduced crop yields (Table 5) in the initial year of the experiment due to cloddy seedbed conditions that reduced crop establishments (Table 4). There was no response to the deep ripping treatments in either the 2017 or 2018 growing season which received the lowest rainfall (319-322 mm). However, in 2016 growing season in the year of highest rainfall (462 mm) the highest grain yield occurred with one of the deep ripping treatments (3.15-3.73 t/ha), irrespective of row spacing (Table 5). This response of deep ripping in wet years is consistent with the findings of Wong and Asseng (2007).

Cereal (wheat and barley) grain yield in 2015, 2016 and 2017 were significantly (p<0.001) higher on narrow row spacing (30 cm) compared to wide row spacing (60 cm), average over mulching and irrespective of treatments (Table 5). For example, in 2016 higher grain yield was observed on narrow row spacing (3.5 t/ha) compare to wide row spacing (2.97 t/ha), average over treatments. This was mainly due to a low number of plants number in wide row spacing compare to narrow row spacing (Table 4). Nevertheless, in 2018 chickpea grain yield has no significant differences in between narrow and wide row spacing, average over treatments.

Grain yield increased in response to gypsum and mulching in the years of lowest rainfall (2017 and 2018, 322 and 319 mm respectively) while no response occurred in the year of highest rainfall (2016, 462 mm). For example, during 2017 and 2018 higher grain yield was found from gypsum treated plots compare to without gypsum, irrespective of mulching (Table 5). However, there was a greater increase in yield in response to gypsum without mulch than with mulch. For example, in 2017 without mulch barley grain yield was increased about 15% and with mulch 4% compared to untreated control, irrespective of row spacing (Figure 1a). This was mainly due to improved soil water infiltration, reduced soil sodicity (ESP), and improved soil Ca:Mg in response to the gypsum.

Figure 1: Effect of mulching on grain yields during 2017 (barley) and 2018 (chickpea), average over row spacing.

On the other hand, during 2017 and 2018 in the years of lowest rainfall (319-322 mm), significant (p<0.001) increases in grain yield were achieved with mulching, irrespective of treatments (Figure 1). For example, in 2018 growing season highest grain yield was found under mulch (1.60 t/ha) than without mulch (1.32 t/ha), average over treatments. This was may be due to the mulching maintained higher soil moisture and reduce surface crusting (Daniells, 2012).

Conclusion

Improvement of crop production on heavy-textured soils with subsoil sodicity is difficult since responses are highly variable from year to year. The results show that narrow row spacing (30 cm) has significant yield benefits over wide row spacing (60 cm), while applying gypsum, deep tillage, mulching and water harvest furrow had some positive impact, but responses were highly variable and depended on seasonal conditions, particularly rainfall. Positive responses to deep ripping were observed in the wetter year but not the drier years. On the other hand, gypsum and mulching responses were observed in the drier year but not the wetter years. Further research is needed to determine the reasons for variable responses of each subsoil constraint and to provide economic evaluation of different ameliorative techniques or combinations of these techniques for soils within each rainfall zone.

References

Blackwell P, Isbister B, Riethmuller G, Barrett-Lennard E, Hall D, Lemon J, Hagan J and Ward P. (2016). Deeper ripping and topsoil slotting to overcome subsoil compaction and other constraints more economically: way to go! 2016 GRDC Grains Research Update, Perth WA.

Daniells IG. (2012). Hardsetting soils: a review. Australian Journal of Soil Research, 50:349-359.

Petersen E. (2016). Economic analysis of the impacts and management of subsoil constraints. 2016 GRDC Grains Research Update, Perth WA.

Schoknecht N and Pathan S. (2013). Soil groups of Western Australia. Resource Management Technical Report 380, Department of Agriculture and Food, Western Australia.

Woang NTF and Asseng S. (2007). Yield and environmental benefits of ameliorating subsoil constrains under variable rainfall in a Mediterranean environment. Plant and Soil, 297:29-42.

Acknowledgments

The research undertaken as part of ‘Subsoil Constraints – understanding and management’ project (DAW00242) is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC, the authors would like to thank them for their continued support. Thanks are due to Merredin RSU for their technical assistance and Mario D’Antuono (DPIRD) for statistical analysis.

GRDC Project Number: DAW00242

Paper reviewed by: Dr James Fisher