Soil Physics
Water Retention, Gas Transport, and Pore Network Complexity during Short-Term Regeneration of Soil Structure Emmanuel Arthur*
Dep. of Agroecology Faculty of Science and Technology Aarhus Univ. Blichers Allé 20 P.O. Box 50 DK-8830 Tjele, Denmark
Per Moldrup
Dep. of Biotechnology, Chemistry and Environmental Engineering Aalborg Univ. Sohngaardsholmsvej 57 DK-9000 Aalborg, Denmark
Per Schjønning Lis Wollesen de Jonge
Dep. of Agroecology Faculty of Science and Technology Aarhus Univ. Blichers Allé 20 P.O. Box 50 DK-8830 Tjele, Denmark
Soil structure maintains prime importance in determining the ability of soils to carry out essential ecosystem functions and services. This study quantified the newly formed structure of 22-mo field-incubated physically disturbed (2-mm sieved) samples of varying clay mineralogy (illite, kaolinite, and smectite) amended with organic material (7.5 Mg ha−1). The newly formed structure was compared with that of sieved, repacked (SR) and natural intact samples described in previous studies. Assessment and comparison of structural complexity and organization was done using water retention (pore size distribution), soil gas diffusivity, air permeability, and derived pore network complexity parameters. Significant decreases in bulk density and increases in pores >100 mm were observed for incubated samples compared with SR samples. For the soils studied, the proportion of pores >100 mm increased in the order: smectite < illite < kaolinite, with no effect of organic amendment. Soil structural complexity, quantified by soil gas diffusivity, air permeability, and derived pore network indices, was greater for incubated than SR samples. For illitic soils, incubated samples had lower water content and higher air-filled porosity and air permeability than natural intact samples at a matric potential of −10 kPa. Despite this, soil pore organization was similar for both natural and incubated soils, but pore network complexity increased in the order: SR < incubated < natural soils. Finally, the air permeability percolation threshold corresponding to the physically based diffusion threshold increased with structural complexity (SR = 0.02 mm2; incubated = 0.20 mm2; natural = 0.70 mm2). Thus, critical reexamination is needed of the often-used 1.0-mm2 percolation threshold for convective air transport when analyzing pore network complexity. Lack of a clear effect of organic amendment for incubated samples suggests using higher application rates in future studies. Abbreviations: OC, organic carbon; PO, pore organization; PSD, pore size distribution; SR, sieved and repacked.
S
oil structure maintains prime importance in determining the ability of soils to carry out essential ecosystem functions and services such as transport and storage of water and gases, turnover of organic matter, provision of optimal conditions for microbial activity, and C sequestration (Lal, 2004; Gregory et al., 2007). Soil structure is quantified by the stability of aggregates, which dictates their resistance to water and wind erosion (Marshall and Quirk, 1950; Chepil, 1951; Le Bissonnais and Arrouays, 1997). Other means of structure quantification include assessment of volume, configuration, and continuity of the pore space, determined by water retention, gas exchange measurements, and computed tomography analyses (Kirkham et al., 1959; Moldrup et al., 2001; Naveed et al., 2013), and level of compactness (Keller and Håkansson, 2010). Soil Sci. Soc. Am. J. 77:1965–1976 doi:10.2136/sssaj2013.07.0270 Received 9 July 2013. *Corresponding author (
[email protected];
[email protected]). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.
Soil Science Society of America Journal
Soil organic matter (SOM) helps to create and stabilize the soil structure through aggregation, and a huge volume of literature explores the possibility of improving, maintaining, or remediating the soil structure by organic amendment (Six et al., 2000; Tejada and Gonzalez, 2007; Clark et al., 2009; Arthur et al., 2011; Eden et al., 2011). Other factors influencing structure formation include wet–dry cycles, freeze–thaw cycles (Pawluk, 1988; Rajaram and Erbach, 1999), and binding and bonding mechanisms arising from biota activity (see review by Degens, 1997). In natural soils, the influence of climatic cycles on soil structure is largely dependent on the rate and magnitude of change in temperature or moisture content (Kay, 1998). Together with other chemical reactions and processes, these climatic cycles are essential for the flocculation of soil particles into aggregates, and these aggregates are then stabilized by the binding and bonding activities of plant roots and microorganisms (Kay, 1998). The type of clay mineral interferes with how organic amendments affect soil structure. Soils rich in highly weathered 1:1 clay minerals (e.g., kaolinite) are less responsive to organic additions and are less susceptible to structural changes following a decrease in SOM contents, whereas soils that are rich in moderately weathered 2:1 clay minerals prove more responsive (Six et al., 2000; Denef and Six, 2005). This phenomenon is attributable to the presence of polyvalent cations, which form complexes with SOM in 2:1 clay-dominated soils. Soils with 1:1 clay minerals, on the other hand, derive part of their stability from the binding capacity of oxides and 1:1 clay minerals (Oades and Waters, 1991). All these factors are expected to play a similar role in the regeneration of structurally disturbed soils (after mining activities, grading and filling, etc.); however, few studies have quantified the gradual recovery or regeneration of soil architecture following structural disturbance (Oyedele et al., 1999; Schjønning et al., 1999; Denef and Six, 2005). Schjønning et al. (1999) used samples from a clay gradient (with similar clay mineralogy) and showed that structurally disturbed temperate soils that had been regenerated during a period of 17 mo had weaker aggregates with a greater volume of large pores relative to their undisturbed counterparts. Oyedele et al. (1999) reported significant aggregation in three tropical soils incubated for 40 d with ground barley (Hordeum vulgare L.) straw/green ryegrass (Lolium perenne L.), with and without disturbance during the incubation period. Denef and Six (2005) also compared the macroaggregation potential of an illitic and a kaolinitic soil and concluded that the effect of added organic matter was prominent in the illitic soil, with physical or electrostatic attractions playing a major role in the kaolinitic soil. Examining the ability of soils with different clay minerals (and similar clay contents) to recover their structure after exposure to similar experimental conditions could provide further insight into the structural regeneration process and the factors involved. To contribute to the very limited data on soil structure regeneration, the primary objectives of the study were to: (i) quantify the newly formed structure of 2-mm sieved soils of varying dominant clay mineralogy (illite, kaolinite, and smectite) after a 22-mo field incubation in lysimeters; (ii) compare this structure 1966
with the structure of sieved and repacked samples; and (iii) assess the role of organic amendments in the process. A secondary objective was to compare the newly formed structure with the structure of natural, intact samples.
MATERIALS AND METHODS Soils and Treatments
Bulk soil was sampled (6–10-cm depth) from seven different fields. Five were Danish moraine soils derived from the Weichselian glacial period (dominated by illite and denoted IL1–IL5), and two (dominated by kaolinite and montmorillonite, denoted KA and MO, respectively) were tropical soils from Ghana. The IL1 and IL2 soils are classified as Glossic Phaeozems, and the IL3, IL4, and IL5 soils are Mollic Luvisols (Krogh and Greve, 1999). The KA soil is an Acrisol (ISSS/ISRIC/ FAO, 1998), and MO is a Calcic Vertisol (FAO/UNESCO, 1990). Details on sampling locations, previous cropping history, and management are provided in Table 1 and Schjønning et al. (2002a). Data for intact natural samples were available in previously published studies for IL1 to IL5 (Schjønning et al., 2002b; Arthur et al., 2012b) and were used for comparison with the present experiments. After sampling, the soils were crushed and sieved to 2 mm. Each soil was separated into two sets. One set of 9 kg of soil from each field was thoroughly mixed with
