Investigation of compounds causing water repellency in the rhizosphere of sandy soils from a wide range of locations

Although soils are generally considered to wet readily, some are actually water repellent at the surface and in the rhizosphere. This phenomenon occurs at low to moderate moisture contents and has been reported from soils under a range of vegetation types and from many regions around the globe. Water repellency in soils can have serious environmental implications including reduced seed germination and plant growth as well as irrigation efficiency, accelerated soil erosion, and enhanced leaching of agrochemicals through preferential flow. It has been proposed that water repellency is caused by the accumulation of hydrophobic organic compounds released as root exudates, microbial byproducts or from decomposing organic matter, which are deposited on mineral or aggregate surfaces, or are present as interstitial matter. Few studies to date have attempted to isolate and characterize these compounds and their structure is therefore only poorly understood. These studies have generally focussed on only a single soil or a small range of samples, have not included non-repellent soils as a control and have not always been able to demonstrate that the substances isolated are indeed responsible for repellency formation. This study reports on the first part (extraction procedures) of a research programme addressing these gaps in current knowledge by investigating a wide range of severely repellent and wettable ‘control’ samples from different countries, and by including assessments of extraction efficiency and ability of extracts to cause repellency. Analytical methods include DRIFT (Diffuse Reflectance Infrared Fourier Transform Spectroscopy) of soils and IR (Infrared) analysis of extracts. Key findings are that (i) soil sample heating after extraction is valuable in assessing the effectiveness of the extraction procedure, (ii) Soxhlet extraction using isopropanol/ ammonia (70/30 v/v) was the most effective method in extracting hydrophobic compounds, while leaving the ability of extracted compounds to induce water DOERR ET AL. 17 WCSS, 14-21 August 2002, Thailand 1520-2 repellency virtually unaffected, (iii) wettable control soils also contain hydrophobic substances capable of inducing water repellency, (iv) the amount of organic compounds extracted was poorly related to sample repellency, indicating that compounds responsible for repellency may only represent a small fraction of the extract, (v) differences in extraction efficiency between different samples indicate that the compounds responsible may differ generically and/or in terms of their bonding to minerals, and (vi) the combination of repellency assessments with DRIFT on soils and IR on extracts used with internal standards has considerable potential to allow quantification of C-H bearing organic matter in the soil, the efficiency of extraction processes for its removal, and its significance in causing water repellency in soils.