Carbon storage in sandy soil amended with clay. Examining the relationship of organic carbon concentration to clay concentration, clod size and distribution

Abstract

Globally, most agricultural systems have lost 40 to 70% of their natural soil organic carbon (OC) through removal of harvest product and past management practice. It is critical to identify and implement practices that minimise or reverse the decline in soil carbon while balancing economic sustainability and global food needs. Increasing OC storage in agricultural soils provides an opportunity to offset greenhouse gas emissions, improve soil health, fertility, structure, water-holding capacity and plant productivity. The amount of stored OC varies among soil types and is strongly influenced by clay concentration. Sandy soils often have low OC content because of low input from limited plant growth and rapid decomposition due to low clay concentration. Sandy soils cover a large proportion of Australia’s agricultural region and are common worldwide. Thus, increasing OC in sandy soils is a good opportunity for increased OC storage. In natural soils, there is a positive correlation between clay and OC concentration because binding of OC to clay reduces decomposition by soil microbes. Therefore, the addition of clay to sandy soil has the potential to increase OC storage. Subsoil clay addition to sandy soils is a practice used in South Australia, Victoria and Western Australia to overcome water repellence and increase water retention, fertility and plant productivity. The addition of subsoil clay to sand creates clods of different sizes, from a few mm to more than 200 mm in diameter distributed in the soil profile. The method chosen for subsoil clay addition can influence clod size and distribution in the soil profile. Little is known about the potential of clay-amended soils to increase OC content and whether clay addition methods can be optimised to increase the OC storage capacity. The aim of the thesis was to determine the effect of subsoil clay addition to sand on clay distribution and OC content. A series of field studies and two incubation experiments were carried out to determine the effect of subsoil clay addition to sand on OC and clay distribution and content. Procedures were validated to ensure sampling captured the variability of OC and bulk density in clay-amended soil. The distribution of clay and OC in the soil profile was quantified for clay-amendment methods. In addition, the effects of clod size, clod chemical properties and amount of added subsoil clay on OC concentrations were evaluated. In the first study the number of soil samples required within a 25-m grid for accurate OC and bulk density measurement in clay-amended soils was determined. Further, OC concentration, bulk density and OC stocks in sandy soil without and with clay addition were measured. The study was carried out on two agricultural properties in South Australia (one in the South East, the other on Eyre Peninsula), where sandy soils without clay addition (1–3% clay) were compared with three methods of kaolinitic subsoil clay addition to sand (clay spread, delved or spaded). Within a 25 m x 25 m sampling area, twenty randomly allocated soil cores to 50 cm depth were collected after harvest. The results showed that 10 randomly allocated cores within a 25 m x 25 m sampling area was sufficient to represent the variability of OC concentration in sandy soil without and with clay addition. Two to three samples were required within the sampling area for accurate representation of bulk density. Stratified sampling is recommended for delved sites with sample allocation based on the proportion of area represented by delve lines and that between delve lines. Subsoil clay addition to sand increased OC stock in 0-30 cm depth by up to 14 t ha ⁻¹ in the South East and 22 t ha⁻¹ on the Eyre Peninsula. OC stock increase was site specific. OC stock was influenced by the clay addition method and dependent on the amount of clay added and depth of incorporation. Clay spreading increased clay and OC close to the soil surface, delving increased them at depth and spading distributed OC and clay evenly within the mixed depth. The second study assessed i) size, number and vertical distribution of clods and OC at two field sites with clay addition and ii) the effect of clod size and properties on OC in incubation experiments. Two field sites, Eyre Peninsula (EP) and South East (SE) with different clay addition method, spaded or delved, were studied. Soil was excavated from a 30 cm quadrat in 10 cm increments from 0 to 40 cm, sieved into various clod sizes in which soil mass, clod number and OC concentration were determined. Subsoil clay properties from 40-60 cm depth including clay concentration, pH, exchangeable cations, iron and clay mineralogy from both sites were analysed. Delving elevated clay and created few, mainly large clods, which were poorly distributed in the sandy soil. Spading mixed clay from 20-30 cm below the soil surface and created many, smaller clods, which were more evenly distributed within 0-30 cm than with delving. OC concentration was highest in the smallest clods, particularly those from close to the soil surface. Clod number per unit of soil mass was more important for OC stock than OC concentration of the clods. Clods collected from the two field sites were further used in incubation experiments to determine the effect of clod size and properties (clay and iron concentration) on potential accumulation and protection of OC. In the accumulation experiment, subsoil clay was collected at depth greater than 40 cm from EP and SE crushed and sieved to clods of 2-6 and 6-20 mm size. The clods were added to sand at 80 mg clay g ⁻¹ sand and incubated with monthly wheat residue addition for 300 days at a water content optimal for microbial activity. In the protection experiment, 2-6 mm and 6-20 mm clods collected from 0-10 cm depth at the SE site were added to sand at 80 mg clay g ⁻¹ sand and incubated for 420 days at a water content optimal for microbial activity. Smaller clods (2-6 mm) accumulated OC at a higher rate and offered greater protection to decomposition by microbes than larger clods (6-20 mm). Clod properties, clay concentration and sesquioxide content influenced OC concentration of 6-20 mm clods but not the 2-6 mm clods. This suggested that the large surface area of the 2-6 mm clods minimised the effect of clod properties. Furthermore, clod number was a critical factor in increasing OC stock. Due to different clay concentration in the EP (58%) and SE (39%) subsoil, to achieve the same target clay addition rate more clods were added per pot for SE compared to EP. OC concentration of the 2-6 mm clods did not differ between EP and SE, thus the higher OC stock in the clods from the SE site was driven by clod number (mass). The higher clod number increased the total surface area and thus the chance that added wheat residue would come in contact with the clod surface. We conclude that in clay-amended soils the addition of many, smaller clods distributed throughout the depth of modification can maximise OC content. The third study aimed to i) compare OC stock in a range of clay-amended and unamended sandy soils under cereal cropping in South Australia and ii) identify factors that influence OC stock to develop best practices to increase OC storage in clay-amended sandy soils. The study was carried out on four agricultural properties in South Australia. Soil OC content, clay content and selected physical and chemical properties of clay-amended treatments and unamended sands were measured. Clay amendment treatments differed in the method of clay addition (clay spread or delved), depth of incorporation (shallow and deep) and amount of subsoil clay added to the surface 30 cm. Soil cores to 50 cm depth were collected within a 25 m x 25 m sampling area. There was a positive correlation between OC and clay stock, but it only explained 46% of the variation in OC stock. This indicated that other factors influenced OC stock in clay-amended treatments. Even vertical distribution of clay within the surface 30 cm was a key factor in increasing OC storage in clay-amended soils. Subsoil clay properties and amount added to the surface 30 cm, as well as depth to undisturbed subsoil also influenced OC stock.