Simultaneous Loss of Soil Biodiversity and Functions

Published August 6, 2014

Soil Physics

Simultaneous Loss of Soil Biodiversity and Functions along a Copper Contamination Gradient: When Soil Goes to Sleep Muhammad Naveed*

Mogens Nicolaisen

Dep. of Agroecology Aarhus Univ. Blichers Allé 20, Postbox 50 DK-8830 Tjele Denmark


Per Moldrup

Markus Tuller

Dep. of Civil Engineering Aalborg Univ. Sohngaardsholmsvej 57 DK-9000 Aalborg Denmark

Dep. of Soil, Water and Environ. Sci. Univ. of Arizona 1177 E. 4th St. Tucson, AZ 85721

Lasantha Herath

Emmanuel Arthur

Dep. of Agroecology Aarhus Univ. Blichers Allé 20, Postbox 50 DK-8830 Tjele


Denmark

Shoichiro Hamamoto

Martin Holmstrup

Graduate School of Agricultural and Life Sciences Univ. of Tokyo 1-1-1, Yayoi Bunkyo-Ku, Tokyo 113-8657 Japan

Dep. of Bioscience Aarhus Univ. Vejlsøvej 25 DK-8600 Silkborg Denmark

Ken Kawamoto Toshiko Komatsu

Dep. of Civil and Environ. Engineering Saitama Univ. 255 Shimo-Okubo Sakura-Ku, Saitama 338-8570 Japan

Hans-Jörg Vogel

Dep. of Soil Physics Helmholtz Center for Environmental Research-UFZ Theodor-Lieser-Straße 4 06120 Halle (Saale) Germany

Lis Wollesen de Jonge Dep. of Agroecology Aarhus Univ. Blichers Allé 20 Postbox 50 DK-8830 Tjele Denmark

The impact of biodiversity loss on soil functions is well established via laboratory experiments that generally consider soil biota groups in isolation from each other, a condition rarely present in field soils. As a result, our knowledge about anthropogenic-induced changes in biodiversity and associated soil functions is limited. We quantified an array of soil biological constituents (plants, earthworms, nematodes, bacteria, and fungi) to explore their interactions and to characterize their influence on various soil functions (habitat for soil organisms, air and water regulation, and recycling of nutrients and organic waste) along a legacy Cu pollution gradient. Increasing Cu concentrations had a detrimental impact on both plant growth and species richness. Belowground soil biota showed similar responses, with their sensitivity to elevated Cu concentrations decreasing in the order: earthworms > bacteria > nematodes > fungi. The observed loss of soil biota adversely affected natural soil bioturbation, aggregate formation and stabilization, and decomposition and mineralization processes and therefore resulted in compacted soil with narrow pore size distributions and overall smaller pores, restricted air and water storage and flow, and impeded C, N, and P cycling. The simultaneous evolution of soil biodiversity and functions along the Cu gradient emphasized the key role of soil life in controlling ecosystem services. Furthermore, results indicated that different soil biodiversity and functional indicators started to decline (10% loss) within a Cu concentration range of 110 to 800 mg total Cu kg−1. Abbreviations: ACE, Abundance Coverage-based Estimator; CT, computed tomography; DHA, dehydrogenase activity; FDA, fluorescein diacetate; SWC, soil water characteristic. Soil Sci. Soc. Am. J. 78:1239–1250 doi:10.2136/sssaj2014.02.0052 Received 4 Feb. 2014. *Corresponding author ([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

S

oils are the biogeochemical engine of the Earth’s life support system and provide vital functions to society, supporting and sustaining our terrestrial ecosystems; growing our food, feed, and fiber; regulating the atmosphere; and providing a vital gene pool and biological resource from which many of our antibiotics have been derived (Robinson et al., 2012; Midgley, 2012). Only a healthy soil ecosystem with balanced biotic and abiotic components interacting in a continuous cycle can provide and sustain these essential life support services. However, the various biotic and abiotic components of soil ecosystems have traditionally been considered in isolation from each other. There is now increasing recognition of the intimate link and feedbacks between biotic and abiotic soil components and their combined effect on ecosystem functions (Young and Crawford, 2004; Wardle et al., 2004). Plants provide organic C for functioning of the decomposer subsystem (organisms that break down organic material, together with their predators) and also enhance natural soil bioturbation, thereby affecting air and water distribution and movement in soils (Gabet et al., 2003; Devitt and Smith, 2002). The decomposer subsystem in turn breaks down dead plant material and indirectly controls plant growth and community composition by regulating the nutrient supply (Wardle et al., 2004). The activity of the decomposer subsystem greatly depends on abiotic boundary conditions such as climate, land use, soil type, and structure because bacteria reside in water-filled micropore spaces (10 mm) (Effmert et al., 2012). Bacteria are normally found on the surfaces of mineral and organic particles, and their spatial distribution is dependent on water movement, while fungi can extend through aerated partially water-saturated pores and spread much farther through the soil (Lavelle and Spain, 2001). Larger organisms have their specific role in the soil ecosystem. Nematodes graze on bacteria and fungi and thus stimulate or balance their growth and distribution and facilitate nutrient cycling by releasing excess NH4 (Irshad et al., 2011). Earthworms derive their nutrients from plant debris and in turn promote microorganism activity by shredding and expanding the surface area of organic matter and also have the ability to steadily rebuild soil structure through their feeding and burrowing activities (Blouin et al., 2013). Lubbers et al. (2013) suggested that although earthworms are largely considered beneficial to soil fertility and plant growth, they increased greenhouse gas emissions and caused a 16% net increase in the global warming potential. Anthropogenic disturbances such as intensive human exploitation, an organism’s habitat disruption and fragmentation, soil pollution, or soil compaction are adversely affecting the delicate balance between biotic and abiotic components of terrestrial ecosystems worldwide ( Jeffery et al., 2010; Hartman et al., 2008). In a recent study encompassing Europe, on 56% of the European Union territory soil biodiversity was under threat to various degrees. The areas under high, very high, and extremely high threat are 9, 4, and 1% of the EU territory, respectively (Gardi et al., 2013). The consequences of declining soil biodi1240

versity have long been of interest for sustainable land use and management and its response to climate change (de Vries et al., 2013). However, most relationships between soil biota and functions have been established by means of controlled laboratory experiments, generally considering soil biota groups and functions in isolation from each other (Nielsen et al., 2011; Philippot et al., 2013; Maestre et al., 2012). The most influential field-scale studies have only focused on the relationships between plant diversity and ecosystem productivity (Midgley, 2012; Tilman et al., 2012), neglecting belowground soil biota and their interactions with the aboveground biomass. Therefore, our knowledge about interactions between soil biotic and abiotic components and their functioning under anthropogenic forcing in the field is limited. Soil is a biologically, physically, and chemically diverse entity that forms the basic substrate of terrestrial ecosystems (Dominati et al., 2010). Therefore, joint evaluation of soil biological (plants, earthworms, nematodes, bacteria, fungi, and microbial activity), physical (soil structure, air and water regulation), and chemical (C, N, and P cycling) characteristics is essential for establishment of the relationships between soil biodiversity and functions in natural ecosystems. We hypothesized that increasing anthropogenic disturbances (in this case, Cu contamination) simultaneously impact various soil biodiversity components and functions. An agricultural field site in Hygum, Denmark, with a legacy Cu contamination gradient provided a unique natural laboratory for testing this hypothesis by exploring: (i) if biodiversity, community composition, and bioactivity changes as a function of soil copper concentration; (ii) if there is evidence for simultaneous loss of soil biodiversity and functions (habitat for soil organisms, air and water regulation, and recycling of nutrients and organic waste) along the Cu contamination gradient; and (iii) if there are critical Cu concentration thresholds that can be linked to detrimental declines in soil biodiversity and functions.

MATERIALS AND METHODS Study Site and Soil Sampling

The research site is a flat, 60- by 120-m rectangular field located in Hygum, Denmark (55°46¢ N, 9°27¢ E). The Cu pollution at the site originated from the application of CuSO4 at a wood preservation plant that was active between 1911 and 1924. The Cu was incorporated into the top 20 cm of soil due to leaching and later tillage operations. The field was cultivated until 1993; thereafter it has been abandoned and left as fallow. Copper is the only contaminant that can be detected at the site. The Cu concentrations follow a gradient that ranges from a background level of 22 mg kg−1 to highly toxic levels of up to 3800 mg kg−1. Five measurement locations were selected along the Cu gradient, three with relatively low Cu concentrations (i.e., 22, 175, and 466 mg kg−1) and two with highly toxic levels (2228 and 3837 mg kg−1) (Fig. 1A). A comprehensive set of measurements (Fig. 1B) was performed at each of the five locations. Earthworm density and plant species richness were measured in triplicate at each location along the Cu gradient. Two representative bulk soil samples (0–10-cm Soil Science Society of America Journal

depth) were sampled in plastic vials from each sampling location and later freeze-dried and homogenized in the laboratory for DNA extraction and sequence processing and analysis. A total of 20 undisturbed soil cores, four from each location, were extracted with aluminum cylinders (10-cm diameter and 8 cm tall). The cores were used for X-ray computed tomography (CT) scanning and for laboratory measurements of soil water characteristics (above −100 kPa matric potential) and gas transport parameters. Twenty small undisturbed soil cores (1 cm3) were then sampled from the larger cores for X-ray micro-CT scanning and for Hg intrusion porosimetry measurements. All intact soil cores were extracted with a customized core sampler from the 5- to 13-cm depth. The soil cores were carefully transported to the laboratory to avoid smearing and compaction effects. In addition, bulk soil samples were collected from the same locations for total and bioavailable Cu (n = 7), pH (n = 7), organic C, P, N, soil texture, specific surface area (n = 1), and soil-water characteristic (below −100 kPa matric potential) measurements (methods are described below).

Soil Biological Indicators

Fig. 1. (A) Link between plant growth and soil pore structure at five measurement locations along the Cu contamination gradient, and (B) a cross-disciplinary framework and associated investigated discipline-specific soil health indicators.

Plant species richness was estimated in triplicate by observing the number of distinct plant species in a 1-m2 area at each location along the Cu gradient. The used nomenclature follows Hansen (1991) for vascular plants and Andersen et al. (1976) for mosses. Earthworms were sampled in spring 2010 by hand sorting of soil cubes covering 0.25 by 0.25 m to a depth of 0.25 m at five locations along the identified gradient. Three cubes were excavated from each sampling location with a spade and placed in plastic buckets. The excavated soil was then spread onsite and the earthworms carefully sorted by hand. All collected worms were carefully kept on moist filter paper in petri dishes for 1 d to allow emptying of the gut for identification. The earthworms were identified at the species level according to the procedure presented by Sims and Gerard (1985). Two samples from each location were freeze-dried and homogenized. The DNA was extracted from a 250-mg subsample using the PowerLyzer PowerSoil DNA Isolation Kit (Mo Bio Laboratories) according to the manufacturer’s instructions. The DNA concentration and quality were determined on a 1% agarose gel. To generate amplicons for pyrosequencing, different primer sets were used: for bacteria, a fragment of the 16S small subunit ribosomal gene was amplified using primers F515 and R806 (Bates et al., 2011); for fungi, the ITS1 region was amplified using primers ITS1-F and 58A2R (Gardes and Bruns, 1993; Martin and Rygiewicz, 2005); and for nematodes, primers NemF (GGGGAAGTATGGTTGCAAA, www.soils.org/publications/sssaj

this study) and 18Sr2b were used (Porazinska et al., 2009). The primers were tag encoded using the forward primer (5¢-CGTATCGCCTCCCTCGCGCCATCAGMID-specific primer-3¢) and the reverse primer (5¢-CTATGCGCCTTGCCAGCCCGCTCAG-specific primer-3¢). Ten-nucleotide multiplex identifier (MID) primer tags for sample identification after pooling were selected from the list of recommended MID primer tags from Eurofins MWG GmbH. Primers were synthesized and highperformance liquid chromatography purified by Eurofins MWG GmbH. Polymerase chain reaction (PCR) contained 1 ´ PCR reaction buffer, 1.5 mmol L−1 MgCl2, 0.2 mmol L−1 deoxynucleotide triphosphates (dNTPs), 1 mmol L−1 each primer, 1 U of Taq DNA recombinant polymerase (Promega Corporation), and 1 mL of DNA template in a final volume of 25 mL. All amplifications were conducted in a GeneAmp PCR System 9700 thermal cycler (PE Applied Biosystems) using an initial DNA denaturation step of 94°C for 5 min, followed by 35 cycles at 94°C for 1 min, 53°C for 1 min, 72°C for 1 min, and finally an elongation at 72°C for 10 min. Integrity of the PCR products was tested by means of electrophoresis in a 1% agarose gel. The concentration of amplicons was estimated by gel electrophoresis and by analysis on a Nanodrop ND 1000 spectrophotometer (Thermo Scientific) 1241

according to the manufacturer’s instructions. Tagged PCR amplicons from each sample were pooled in approximately equimolar amounts to obtain three pools (bacterial, fungal, and nematode). The pooled amplicons were electrophoresed in 1.5% agarose gel, and the smear of PCR products at the relevant size was cut from the gel and purified using a QIAquick Gel Extraction Kit (Qiagen GmbH). The pool was sequenced by Eurofins MWG on a GS Junior 454 Sequencer (Roche Diagnostics). After an initial quality filtering at Eurofins MWG, sequences that were shorter than expected, sequences that contained ambiguities, and sequences where primers and tags did not match were excluded from the analysis. Fungal sequences were clustered at the PlutoF platform (http://elurikkus.ut.ee/plutof. php?lang=eng), bacterial sequences were clustered and analyzed using a combination of the Vamps (http://vamps.mbl.edu/ resources/software.php) and the Scata (http://scata.mykopat. slu.se) platforms, and nematode sequences were analyzed with the CLOTU platform (http://www.lifeportal.uio.no), all set at 97% identity thresholds for clustering. Singletons were omitted from further analysis. Individual operational taxonomic units were identified by Blast searching the National Center for Biotechnology Information. Non-nematode sequences in the nematode data set were filtered. The Abundance Coverage-based Estimator (ACE) richness and Shannon–Wiener diversity indices were calculated using EstimateS (http://viceroy.eeb.uconn. edu/estimates). A total of 23,608 non-singleton sequences from fungi (1511–7208 in each sample), 30,416 sequences from bacteria (1157–3324 in each sample), and 33,179 sequences from nematodes (732–7682 in each sample) were obtained. Microbial activity was estimated for air-dried aggregates in triplicate using two methods: fluorescein diacetate (FDA) (3¢,6¢-diacetylfluorescein) hydrolysis and dehydrogenase activity (DHA) by iodonitrotetrazolium reduction. Fluorescein diacetate hydrolysis activity was determined as described by Schnürer and Rosswall (1982). Dehydrogenase activity was determined by reduction of 2-p-iodo-nitrophenyl-phenyltetrazolium chloride (INT) to iodo-nitrophenyl formazan (INTF) following the method of García et al. (1993).

Soil Physical Indicators Soil texture was determined for 2-mm sieved soil samples with a combination of wet sieving and hydrometer methods (Gee and Or, 2002). Soil specific surface area was measured by means of the ethylene glycol monoethyl ether (EGME) method (Pennell, 2002). The soil water characteristic (SWC) measurements were performed in the laboratory at constant temperature (20°C). The soil cores were placed in a sand box and saturated with water from the bottom. After saturation, suction was successively applied to establish matric potentials of −3 and −10 kPa. The Richard’s pressure plate apparatus (Tuller and Or, 2005) was used for measuring the SWC at −30 and −100 kPa matric potentials. The dry part of the SWC was measured on small bulk samples with a WP4-T dewpoint potentiameter (Decagon Devices). 1242

An industrial X-ray micro-CT scanner (X-Tek HMX225) at the Helmholtz Center for Environmental Research in Halle, Germany, was used to scan the large soil samples (10-cm diameter and 8 cm tall) at an energy level of 200 kV and 500 mA with 800 angular projections. A Cu filter of 0.5-mm thickness was positioned between the X-ray source and the samples to alleviate beam hardening. The obtained shadow projections (radiographs) were reconstructed with a Feldkamp cone beam algorithm (Feldkamp et al., 1984) to obtain 16-bit, grayscale, three-dimensional volumes composed of 500 by 500 by 400 voxels (resolution = 200 mm). An industrial X-ray microCT scanner, HMX225-ACIS (Tesco Corporation) at Saitama University, Japan, was used to scan the 1-cm3 soil samples. The smaller samples were scanned at an energy level of 180 kV and 200 mA with 1600 angular projections yielding a three-dimensional data set of 1024 by 1024 by 1024 voxels at a resolution of 10 mm. The lower resolution data sets (from the large cores) were used to derive the pore size distributions (>300 mm), while the high-resolution data (from the small cores) provided the distributions for pore sizes >10 and £300 mm by excluding the pores >300 mm. The 3DMA-Rock CT data analysis software package (Lindquist, 2010) was used to extract pore size distributions. Quantitative investigation of the pore space features first requires a transformation of fuzzy grayscale data into the porous media phases of interest (in this case, a binarization of grayscales into pore space and the solid soil matrix). The indicator kriging algorithm that is part of 3DMA-Rock was used for the segmentation process. This can be done by manually establishing a window of intensity values delimited by two thresholds. Intensity values above the upper threshold (16,000) are associated with the matrix and those below the lower threshold (10,000) are associated with the void space. The classification of the remaining voxels (between the thresholds) is accomplished with indicator kriging, which utilizes an estimate of a correlation function incorporating local spatial information. Geometric analysis of a three-dimensional, irregularly shaped object such as a soil pore is challenging. Therefore skeletonization of the soil pore system (medial axis construction) was performed using an algorithm by Lee et al. (1994) that provides a representation of the pore objects of interest at a lower dimension, which are easier to analyze (Lindquist and Venkatarangan, 1999). The constructed medial axes were then used to determine the locations of throats in the pore channels (Lindquist and Venkatarangan, 1999). In a manner analogous to the erosion procedure used to reduce the pore channel to its medial axis, we uniformly dilated each medial axis segment in the radial direction perpendicular to its length so that it became a solid cylinder. At some point this continued dilation was halted, and a closed loop of contacting grain points encircling the cylinder was formed. By definition, this closed loop represented the throat of the analyzed pore segment. Based on these located so-called pore throats, the pore system was then separated into individual pores (Lindquist and Venkatarangan, Soil Science Society of America Journal

1999). The size of each nodal pore was measured as the diameter of the equivalent sphere separated by throat surfaces. Four intact soil cores (1 cm3) from each location along the Cu gradient were used for measuring porosity and pore size distribution with Hg intrusion porosimetry (ASTM, 2010). Air permeability (ka) was measured with the steady-state method described by Iversen et al. (2001) at −10 kPa matric potential. The pressure gradient was established at 0.5 kPa to ensure laminar flow during the measurements. The ka was calculated from Darcy’s equation based on the pressure difference across the core. Gas diffusivity (DP/D0) was also measured under the same moisture and matric potential conditions by means of a non-steadystate method using a two gas–dual chamber gas diffusion measurement device (Schjønning et al., 2013). This device consists of two chambers purged with different tracer gases, O2 and Ar. The method involves frequent measurements of both the change in the O2 concentration in the chamber previously flushed with Ar and the change in the Ar concentration in the O2 chamber.

Soil Chemical Indicators Atomic absorption spectrometry (AAS) with flame atomizer was used for soil Cu analysis. Air-dried and sieved (2-mm) soil of 0.3 g mass was crushed in a mortar. Two milliliters of 7 mol L−1 HNO3 (Merck) was added, followed by heating to 80°C for 17 h, and finally evaporation of the fluid at 110°C until dryness. Another 1 mL of 7 mol L−1 HNO3 was added to each sample and the procedure was repeated. The samples were redissolved in 5 mL of 0.1 mol L−1 HCl and then analyzed with AAS (PerkinElmer 4100). A certified reference soil (VKIJ1, Danish Hydraulic Institute) was analyzed as a standard. Extraction with 0.01 mol L−1 CaCl2 was used as a measure of the available Cu fraction. Twenty milliliters of 0.01 mol L−1 CaCl2 was added to 2 g of dry soil, and the sample was shaken end-over-end for 20 h and then centrifuged at 5000 rpm for 5 min at room temperature (Fischer Scientific). The supernatant was used for Cu analysis with AAS as previously described. In total, seven samples from each location were analyzed for total and CaCl2–available Cu (Rhoades, 1982). Soil pH was measured in a solution of 8 mL of air-dry soil and 30 mL of deionized water. The mixture was mechanically shaken for 10 min and left to settle for another 10 min. An H+ ion-selective electrode was then used to measure the soil pH. Soil organic C was determined on pulverized samples via oxidation of C to CO2 at 1800°C with a FLASH 2000 organic elemental analyzer coupled to a thermal conductivity detector (Thermo Fisher Scientific) (Nelson and Sommers, 1996). Total N was determined by means of dry combustion “elemental analysis” (ISO, 1998). Phosphorus was extracted by shaking 1 g of air-dry soil dispersed in 10 mL of 0.025 mol L−1 HCl and 0.03 mol L−1 NH4F for 5 min. Phosphorus was determined on the supernatant with the molybdenum-blue method using ascorbic acid as a reductant. Color development was measured at 880 nm with a Brinkmann PC 800 probe colorimeter (Metrohm USA). www.soils.org/publications/sssaj

Applied Physical Models The Campbell (1974) soil water retention model parameter (b) was obtained as the slope of the linear regression fitted to measured SWC data between −1 and −100 kPa matric potentials:

y  q = qs  e   y 

1/b

[1]

where ye (hPa) is the soil water potential at air entry, q (m3 m−3) is the volumetric soil water content, qs (m3 m−3) is the soil water content at saturation, and b (dimensionless) is an empirical constant. The Rosin–Rammler model (Rosin and Rammler, 1933) was used to describe the pore size distributions:

   x  b*   P ( X < x=) 100 1− exp  −       a *   

[2]

where P(X < x) is the percentage of pores less than size x, and a* and b* are adjustable parameters related to the characteristic pore size (mm) at which 63.2% of the pores are finer and the spread of the pore size distributions, respectively. The free model parameters a* and b* were determined via nonlinear regression analysis using the Microsoft Excel Solver tool. A four-parameter logistic model was used to describe the Cu concentration vs. response data (measured soil biological, physical, and chemical indicators):

= Y Min +

Max − Min

1+ ( X MC )

−H

[3]

where Min and Max are the responses of the parameter corresponding to the minimum and maximum Cu concentrations, MC is the median Cu concentration that can cause a defined effect of 50%, and H is the hillslope that characterizes the slope of the curve at its midpoint.

Statistical Analysis The four-parameter logistic model fitting to Cu concentration vs. response data, correlation analysis, and other regression analyses were performed with the commercial software SigmaPlot 11.0 (Systat Software).

RESULTS AND DISCUSSION The soil texture of the field was classified as sandy loam, with clay, silt, and sand contents ranging from 11 to 12, 22 to 31, and 52 to 67%, respectively. Soil organic matter increased from 3.3 to 6.0% with increasing Cu concentrations (22–3837 mg kg−1), indicating that decomposition of dead biomaterial at highly toxic levels is considerably inhibited (Miltner et al., 2012), as the study site was exposed to Cu pollution over a time scale of 100 yr. Soil specific surface area seamlessly followed organic matter content, increasing from 31 to 40 m2 g−1. The average soil pH(H2O) was about 6.2 and showed no significant variations along the Cu 1243

contamination gradient. The CaCl2–extractable Cu (bioavailable) fraction was linearly correlated to the total Cu concentration (R2 = 0.99) and therefore later was used as a contamination metric (Table 1).

Biodiversity, Community Composition, and Bioactivity Changes as a Function of Copper Concentration A comprehensive set of bioindicators, including plants and belowground soil biota ranging from single-cell microorganisms to nematodes that occupy water films on the surface of soil aggregates, to larger soil inhabitants such as earthworms, was chosen to assess the state of soil biodiversity at the five selected locations. Copper is considered an essential plant micronutrient, but its elevated concentrations in soil can be toxic. In the studied field, which was kept fallow since 1993, the vegetation was visibly affected by increasing Cu content, with an area of a few square meters virtually without vegetation at the highest Cu concentration of 3837 mg kg−1 (Fig. 1A). Plant species richness generally decreased with increasing Cu concentrations, although the highest plant species richness occurred at intermediate Cu concentration (Fig. 2A). Plant community composition at soil Cu concentrations >200 mg kg−1 differed considerably from the community composition at lower Cu levels. Grasses, mosses, and two low-growing plants (Tussilago farfara L. and Equisetum arvense L.) were the most tolerant to elevated Cu concentrations, while Cirsium arvense (L.) Scop., Anthriscus sylvestris (L.) Hoffm., Tanacetum vulgare L., Ulmus glabra Huds., and Epilobium hirsutum L. were among the most sensi- Fig. 2. Soil biological, physical, and chemical indicators plotted as a function of soil tive plants. A few tree species, such as Urtica dio- Cu concentration: (A) plant species richness and earthworm density, (B) nematode Abundance Coverage-based Estimator (ACE) richness and Shannon–Wiener diversity, ica L., Epilobium montanum L. and Ulmus glabra (C) bacterial ACE richness and Shannon–Wiener diversity, (D) fungal ACE richness and Huds., even started colonizing at intermediate Cu Shannon–Wiener diversity, (E) microbial activity indicators fluorescein diacetate (FDA) concentrations (Fig. 3A). Such differences in plant and dehydrogenase (DHA) assays, (F) total porosity and Campbell model parameter b, (G) Rosin–Rammler model parameters a* and b*, (H) gas diffusivity and air permeability tolerance can be caused by differences in their Cu at −10 kPa matric potential, (I) organic C, and (J) total P and N. uptake and excretion rates and a combination of physiological, biochemical, and molecular changes that can imal., 2006). A few other field studies investigating the effects of prove living conditions under heavy metal stress (Strandberg et heavy metal contamination on vegetation found significant imTable 1. Soil physicochemical properties at five measurement locations along the Cu contamination gradient. Location

Total Cu

Bioavailable Cu

———— mg kg−1 ———— 1 22 ± 1† 0.17 ± 0.01 2 175 ± 6 1.64 ± 0.11 3 466 ± 18 5.27 ± 0.27 4 2228 ± 83 26.42 ± 2.23 5 3837 ± 158 49.50 ± 4.45 † Mean values ± standard error (n = 7). 1244

Clay 300-mm diameter) were observed, which may be formed due to freeze–thaw and wet–dry cycles (Fig. 6). Overall, Soil Science Society of America Journal

bell-shaped pore size distributions were observed along the Cu gradient (Fig. 5B–5C). The fraction of soil pores >10 mm in diameter decreased with increasing Cu concentration. This was most likely caused by the loss of microbial activity responsible for the formation of aggregated soil structure due to production of polysaccharides and other cellular debris (Degens, 1997). The fraction of micropores (300 mm) for four soil column 466, 2228, and 3837 mg kg−1 from top to bottom, respectively. Among the soil physical indicators, spread of the soil pore size distribution (Rosin–Rammler b*) and gas diffusivity were most affected by increasing Cu concentrations, nization, where the creation and organization of the pore-scale and their SF10% values were at 170 and 200 mg kg−1, respecarchitecture arise spontaneously due to interactions between tively. Soil chemical indicators (organic C, total N and P) were microbes and material (water, gas, and chemical) fluxes. At the also considerably affected, and their SF10% Cu concentrations control location (22 mg total Cu kg−1), the soil exhibited good −1 lie in the range of 225 to 470 mg kg (Fig. 7). This corroboplant growth and high microbial richness, diversity, and activrates well with the results of Yang et al. (2002). They found soil Cu thresholds for Table 2. Pearson’s correlation coefficients between the soil biodiversity indicators plant species richness (PSR), earthworm density (ED), nematode richness (NR), nematode 10% yield reduction within the range of diversity (ND), bacterial richness (BR), bacterial diversity (BD), fungal richness (FR), 161 to 835 mg kg−1 and potential dietary and fungal diversity (FD) and the functional indicators total porosity (f), Campbell’s toxicity in edible parts of the vegetables soil water characteristic model parameter b, Rosin–Rammler model parameters a* and b*, air-filled porosity at −10 kPa (e), air permeability at −10 kPa (ka), gas diffusivity within the range of 142 to 232 mg kg−1. at −10 kPa (D /D ), organic C (OC), N, and P along the Cu contamination gradient. P 0 Stobrawa and Lorenc-Plucinska (2008) Parameter PSR ED NR ND BR BD FR FD established Cu thresholds resulting in 0.88* 0.95* 0.98** 0.99** 0.96** 0.89* 0.83 0.58 f increasing toxicity for plants within the b 0.79 0.91* 0.93* 0.97** 0.91* 0.89* 0.97** 0.82 range of 220 to 650 mg kg−1. 0.70 0.97** 0.96** 0.94* 0.96** 0.97* 0.87** 0.68 a*

CONCLUSIONS Combining emerging with conventional biophysicochemical measurement techniques enabled a comprehensive evaluation of changes in soil ecosystem health. The results are in agreement with a proposed mechanism of soil self-orga1248

b* e ka

0.65 0.82

0.99** 0.97**

0.96** 0.99**

0.89* 0.97**

0.91* 0.92* 0.97** 1.00** DP/D0 0.81 0.99** 0.99** 0.98** OC −0.80 −0.95* −0.97** −0.97** N −0.62 −0.85* −0.85* −0.74 P −0.72 −0.99** −0.97** −0.94* * Significant correlation at the 0.05 probability level. ** Significant correlation at the 0.01 probability level.

0.96** 0.93*

0.99** 0.97*

0.75 0.86*

0.51 0.64

0.91* 0.88* 0.98** 0.95* −0.92** −0.95** −0.74 −0.92* −1.00** −0.95*

0.91* 0.86* −0.92* −0.55 −0.78

0.70 0.62 −0.74 −0.31 −0.52


Fig. 7. Threshold Cu concentrations indicative of a 10% loss of various biological, physical, and chemical soil indicators (SF10%).

ity, along with a healthy aggregated soil structure with wider soil pore size distribution, higher total porosity, air permeability, and gas diffusivity, and better nutrient cycling. With increasing Cu pollution (22–3837 mg kg−1) reduced plant growth and microbial richness, diversity, and activity yielded unhealthy and increasingly single-grained soil structure. This resulted in narrow soil pore size distributions, lower total porosity, air permeability, and gas diffusivity, and impeded nutrient cycling. The intimately linked evolution of soil biodiversity and functional indicators suggests the need for a new paradigm for exploratory soil science, with broad, cross-disciplinary studies providing the key to understanding the simultaneous development of soil life-support architecture, soil functions, and life and diversity in soils.

ACKNOWLEDGMENTS

This study was part of the large framework project Soil Infrastructure, Interfaces, and Translocation Processes in Inner Space (“Soil-it-is”) funded by the Danish Research Council for Technology and Production Sciences and was also partially supported by the Core Research for Evolutionary Science and Technology (CREST) program of the Japan Science and Technology Agency ( JST).

REFERENCES

Andersen, A.G., D.F. Boesen, K. Holmen, N. Jacobsen, J. Lewinsky, G. Mogensen, et al. 1976. Den danske mosflora: I. Bladmosser. Gyldendalske Boghandel, Copenhagen. ASTM. 2010. D4404-10: Standard test method for determination of pore volume and pore volume distribution of soil and rock by mercury intrusion porosimetry. ASTM Int., West Conshohocken, PA. doi:10.1520/D4404-10 Bates, S.T., D. Berg-Lyons, J.G. Caporaso, W.A. Walters, R. Knight, and N. Fierer. 2011. Examining the global distribution of dominant archael populations in soil. ISME J. 5:908–917. doi:10.1038/ismej.2010.171 Berg, J., K.K. Brandt, W.A. Al-Soud, P.E. Holm, L.H. Hansen, S.J. Sørensen, and O. Nybroe. 2012. Selection of Cu-tolerant bacterial communities with altered composition, but unaltered richness, via long-term copper exposure. Appl. Environ. Microbiol. 78:7438–7446. doi:10.1128/AEM.01071-12 Blouin, M., M.E. Hodson, E.A. Delgado, G. Baker, L. Brussard, K.R. Butt, et al. 2013. A review of earthworm impact on soil function and ecosystem services. Eur. J. Soil Sci. 64:161–182. doi:10.1111/ejss.12025 Brown, G. 1994. Soil factors affecting patchiness in community composition of

www.soils.org/publications/sssaj

heavy metal-contaminated areas of Western Europe. Vegetatio 115:77–90. Campbell, G.S. 1974. Simple method for determining unsaturated conductivity from moisture retention data. Soil Sci. 117:311–314. doi:10.1097/00010694-197406000-00001 Chu, G., S.A. Wakelin, L. Condron, and A. Stewart. 2010. Effect of soil copper on the response of soil fungal communities to the addition of plant residues. Pedobiologia 53:353–359. doi:10.1016/j.pedobi.2010.04.002 de Boer, T.E., N. Tas, M. Braster, E.J. Temminghoff, W.F. Roling, and D. Roelofs. 2012. The influence of long-term copper contaminated agricultural soil at different pH levels on microbial communities and springtail transcriptional regulation. Environ. Sci. Technol. 46:60–68. doi:10.1021/es2013598 Degens, B.P. 1997. Macro-aggregation of soils by biological bonding and binding mechanisms and factors affecting these: A review. Aust. J. Soil Res. 35:431–459. doi:10.1071/S96016 de Jonge, L.W., P. Moldrup, and P. Schjønning. 2009. Soil infrastructure, interfaces & translocation processes in inner space (“Soil-it-is”): Towards a road map for the constraints and cross-roads of soil architecture and biophysical processes. Hydrol. Earth Syst. Sci. 13:1485–1502. doi:10.5194/hess-13-1485-2009 Devitt, D.A., and S.D. Smith. 2002. Root channel macropores enhance downward movement of water in a Mojave Desert ecosystem. J. Arid Environ. 50:99–108. doi:10.1006/jare.2001.0853 de Vries, F.T., E. Thébault, M. Liirie, K. Birkhofer, M.A. Tsiafouli, L. Bjørnlund, et al. 2013. Soil food web properties explain ecosystem services across European land use systems. Proc. Natl. Acad. Sci. 110:14296–14301. doi:10.1073/pnas.1305198110 Dominati, E., M. Patterson, and A. Mackay. 2010. A framework for classifying and quantifying the natural capital and ecosystem services of soils. Ecol. Econ. 69:1858–1868. doi:10.1016/j.ecolecon.2010.05.002 Effmert, U., J. Kalderas, R. Warnke, and B. Piechulla. 2012. Volatile mediated interactions between bacteria and fungi in the soil. J. Chem. Ecol. 38:665– 703. doi:10.1007/s10886-012-0135-5 Ekschmitt, K., and G.W. Korthals. 2006. Nematodes as sentinels of heavy metals and organic toxicants in the soil. J. Nematol. 38:13–19. Feldkamp, L.A., L.C. Davis, and J.W. Krees. 1984. Practical cone-beam algorithm. J. Opt. Soc. Am. 1:612–619. doi:10.1364/JOSAA.1.000612 Fisker, K.V., M. Holmstrup, and J.G. Sørensen. 2013. Variation in metallothionein gene expression is associated with adaptation to copper in the earthworm Dendrobaena octaedra. Comp. Biochem. Physiol., Part C 157:220–226. Gabet, E.J., O.J. Reichman, and E.W. Seabloom. 2003. The effects of bioturbation on soil processes and sediment transport. Annu. Rev. Earth Planet. Sci. 31:249–273. doi:10.1146/annurev.earth.31.100901.141314 Galbraith, H., K. LeJeune, and J. Lipton. 1995. Metal and arsenic impacts to soils, vegetation communities and wildlife habitat in southwest Montana uplands contaminated by smelter emissions: I. Field evaluation. Environ. Toxicol. Chem. 14:1895–1903. doi:10.1002/etc.5620141111 García, C., T. Hernández, F. Costa, B. Ceccanti, and G. Masciandaro. 1993. The dehydrogenase activity of soil as an ecological marker in processes of perturbed system regeneration. In: J. Gallardo-Lancho, editor, Proceedings of the XI International Symposium on Environmental Biochemistry, Salamanca, Spain. CSIC, Salamanca. p. 89–100. Gardes, M., and T.D. Bruns. 1993. ITS primers with enhanced specificity for basidiomycetes: Application to the identification of mycorrhizae and rusts. Mol. Ecol. 2:113–118. doi:10.1111/j.1365-294X.1993.tb00005.x Gardi, C., S. Jeffery, and A. Saltelli. 2013. An estimate of potential threats levels to soil biodiversity in EU. Global Change Biol. 19:1538–1548. doi:10.1111/gcb.12159 Gee, G.W., and D. Or. 2002. Particle size analysis. In: J.H. Dane and G.C. Topp, editors, Methods of soil analysis. Part 4. Physical methods. SSSA Book Ser. 5. SSSA, Madison, WI. p. 255–293. Hansen, K. 1991. Dansk feltflora. Gyldendalske Boghandel, Copenhagen. Hartman, W.H., J.R. Curtis, V. Rytas, and L.B. Gregory. 2008. Environmental and anthropogenic controls over bacterial communities in wetland soils. Proc. Natl. Acad. Sci. 105:17842–17847. doi:10.1073/pnas.0808254105 Holmstrup, M., and H.D. Hornum. 2012. Earthworm colonisation of abandoned arable soil polluted by copper. Pedobiologia 55:63–65. doi:10.1016/j.pedobi.2011.08.005 Irshad, U., C. Villenave, A. Brauman, and C. Plassard. 2011. Grazing by nematodes on rhizosphere bacteria enhances nitrate and phosphorous availability to Pinus pinaster seedlings. Soil Biol. Biochem. 43:2121–2126.

1249

doi:10.1016/j.soilbio.2011.06.015 ISO. 1998. Method 13878: Soil quality: Determination of total nitrogen content by dry combustion (“elemental analysis”). ISO, Geneva, Switzerland. http://www.iso.org/iso/. Iversen, B.V., P. Schjønning, T.G. Poulsen, and P. Moldrup. 2001. In site, on-site, and laboratory measurements of soil air permeability: Boundary conditions and measurement scale. Soil Sci. 166:97–106. doi:10.1097/00010694-200102000-00003 Janvier, C., F. Villeneuve, C. Alabouvette, E. Hermann, M. Thierry, and C. Steinberg. 2007. Soil health through soil disease suppression: Which strategy from descriptors to indicators? Soil Biol. Biochem. 9:957–961. Jeffery, S., C. Gardi, A. Jones, L. Montanarella, L. Marmo, L. Miko, et al. 2010 The European atlas of soil biodiversity. Publ. Office Eur. Union, Luxembourg. Lavelle, P., and A.V. Spain. 2001. Soil ecology. Kluwer Acad. Publ., Dordrecht, the Netherlands. Lee, T.C., R.L. Kashyap, and C.N. Chu. 1994. Building skeleton models via 3-D medial surface axis thinning algorithms. CVGIP: Graph. Models Image Process. 56:462–478. doi:10.1006/cgip.1994.1042 Li, Q., Y. Jiang, and W.J. Liang. 2006. Effect of heavy metals on soil nematode communities in the vicinity of a metallurgical factory. J. Environ. Sci. 18:323–328. Lindquist, W.B. 2010. 3DMA-Rock, a software package for automated analysis of rock pore structure in 3D computed microtomography images. http:// www.ams.sunysb.edu/~lindquis/3dma/3dma_rock/3dma_rock.html (accessed 31 Jan. 2010). Lindquist, W.B., and A.B. Venkatarangan. 1999. Investigating 3D geometry of porous media from high resolution images. Phys. Chem. Earth 24:593– 599. doi:10.1016/S1464-1895(99)00085-X Lubbers, I.M., K.J. van Groenigen, S.J. Fonte, J. Six, L. Brussaard, and J.W. van Groenigen. 2013. Greenhouse-gas emissions from soils increased by earthworms. Nat. Clim. Change 3:187–194. doi:10.1038/nclimate1692 Maestre, F.T., J.L. Quero, N.J. Gotelli, A. Escudero, V. Ochoa, M. Delgado-Baquerizo, et al. 2012. Biodiversity enhances ecosystem multifunctionality in the world’s drylands. Science 335:214–217. doi:10.1126/science.1215442 Martin, K.J., and P.T. Rygiewicz. 2005. Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts. BMC Microbiol. 5:28. doi:10.1186/1471-2180-5-28 Mertens, J., S.A. Wakelin, K. Broos, M.J. McLaughlin, and E. Smolders. 2010. Extent of copper tolerance and consequences for functional stability of the ammonia-oxidizing community in long-term copper contaminated soils. Environ. Toxicol. Chem. 29:27–37. doi:10.1002/etc.16 Midgley, G.F. 2012. Biodiversity and ecosystem function. Science 335:174–175. doi:10.1126/science.1217245 Miltner, A., P. Bombach, B. Schmidt-Brucken, and M. Kästner. 2012. SOM genesis: Microbial biomass as a significant source. Biogeochemistry 111:41–51. doi:10.1007/s10533-011-9658-z Neher, D.A., K.N. Easterling, D. Fiscus, and C.L. Campbell. 1998. Comparison of nematode communities in agricultural soils of North Carolina and Nebraska. Ecol. Appl. 8:213–223. doi:10.1890/1051-0761(1998)008[0213:CONCIA]2.0.CO;2 Nelson, D.W., and I.E. Sommers. 1996. Total carbon, organic carbon, and organic matter. In: D.L. Sparks, editor, Methods of soil analysis. Part 3. Chemical methods. SSSA Book Ser. 5. SSSA and ASA, Madison, WI. p. 961–1010. Nielsen, U.N., E. Ayres, D.H. Wall, and R.D. Bardgett. 2011. Soil biodiversity and carbon cycling: A review and synthesis of studies examining diversity–function relationships. Eur. J. Soil Sci. 62:105–116. doi:10.1111/j.1365-2389.2010.01314.x Pennell, K.D. 2002. Specific surface area. In: J.H. Dane and G.C. Topp, editors, Methods of soil analysis. Part 4. SSSA Book Ser. 5. SSSA and ASA, Madison, WI. p. 308–313. Philippot, L., A. Spor, C. Henault, D. Bru, F. Bizouard, C.M. Jones, et al. 2013. Loss in microbial diversity affects nitrogen cycling in soil. ISME J. 7:1609– 1619. doi:10.1038/ismej.2013.34 Porazinska, D.L., R.M. Giblin-Davis, L. Faller, W. Farmerie, N. Kanzaki, K. Morris, et al. 2009. Evaluating high throughput sequencing as a method for

1250

metagenomic analysis of nematode diversity. Mol. Ecol. Resour. 9:1439– 1450. doi:10.1111/j.1755-0998.2009.02611.x Rajapaksha, R.M.C.P., M.A. Tobor-Kaplon, and E. Baath. 2004. Metal toxicity affects fungal and bacterial activities in soil differently. Appl. Environ. Microbiol. 70:2966–2973. doi:10.1128/AEM.70.5.2966-2973.2004 Rhoades, J.D. 1982. Cation exchange capacity. In: A.L. Page et al., editors, Methods of soil analysis. Part 2. Chemical and microbiological properties. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI. p. 149–157. Robinson, D.A., N. Hockley, E. Dominati, I. Lebron, K.M. Scow, B. Reynolds, et al. 2012. Natural capital, ecosystem services, and soil change: Why soil science must embrace an ecosystems approach. Vadose Zone J. 11(1). doi:10.2136/vzj2011.0051 Rosin, P., and E. Rammler. 1933. Laws governing the fineness of powdered coal. J. Inst. Fuel 7:29–36. Rutgers, M., A.J. Schouten, J. Bloem, N.V. Eekeren, R.G.M. de Goede, G.A.J.M. Jagers op Akkerhuis, and A. van der Wal. 2009. Biological measurements in a nationwide soil monitoring network. Eur. J. Soil Sci. 60:820–832. doi:10.1111/j.1365-2389.2009.01163.x Salamun, P., M. Renco, E. Kucanova, T. Brazova, I. Papajova, D. Miklisova, and V. Hanzelova. 2012. Nematodes as bioindicators of soil degradation due to heavy metals. Ecotoxicology 21:2319–2330. doi:10.1007/s10646-012-0988-y Sauvé, S. 2006. Copper inhibition of soil organic matter decomposition in a seventy-year field exposure. Environ. Toxicol. Chem. 25:854–857. doi:10.1897/04-575R.1 Schjønning, P., M. Eden, P. Moldrup, and L.W. de Jonge. 2013. Two-chamber, two-gas and one-chamber, one-gas methods for measuring the soil-gas diffusion coefficient: Validation and inter-calibration. Soil Sci. Soc. Am. J. 77:729–744. doi:10.2136/sssaj2012.0379 Schnürer, J., and T. Rosswall. 1982. Fluorescein diacetate (FDA) hydrolysis as a measure of total microbial activity in soil and litter. Appl. Environ. Microbiol. 43:1256–1261. Schutter, M.E., J.M. Sandero, and R.P. Dick. 2001. Seasonal, soil type, and alternative management influences on microbial communities of vegetable cropping systems. Biol. Fertil. Soils 34:397–410. doi:10.1007/s00374-001-0423-7 Sims, R., and B. Gerard. 1985. Synopses of the British fauna. Vol. 31. Earthworms. Field Studies Council, Shrewsbury, UK. Stobrawa, K., and G. Lorenc-Plucinska. 2008. Thresholds of heavy-metal toxicity in cuttings of European black poplar (Populus nigra L.) determined according to antioxidant status of fine roots and morphometrical disorders. Sci. Total Environ. 390:86–96. doi:10.1016/j.scitotenv.2007.09.024 Strandberg, B.J., M. Axelsen, J.J. Pedersen, and M. Attrill. 2006. Effect of a copper gradient on plant community structure. Environ. Toxicol. Chem. 25:743–753. doi:10.1897/04-582R.1 Tilman, D., P.B. Reich, and F. Isbell. 2012. Biodiversity impacts ecosystem productivity as much as resources, disturbance, or herbivory. Proc. Natl. Acad. Sci. 109:10394–10397. doi:10.1073/pnas.1208240109 Tuller, M., and D. Or. 2005. Water films and scaling of soil characteristic curves at low water contents. Water Resour. Res. 41:W09403. doi:10.1029/2005WR004142 Wakelin, S.A., G. Chu, R. Lardner, Y. Liang, and M. McLaughlin. 2010. A single application of Cu to field soil has long-term effects on bacterial community structure, diversity, and soil processes. Pedobiologia 53:149– 158. doi:10.1016/j.pedobi.2009.09.002 Wardle, D.A., R.D. Bardgett, J.N. Klironomos, H. Setälä, W.H. van der Putten, and D.H. Hall. 2004. Ecological linkages between aboveground and belowground biota. Science 304:1629–1633. doi:10.1126/science.1094875 Yang, X.-E, X.X. Long, W.-Z. Ni, Z.-Q. Ye, Z.-L. He, P.J. Stoffella, and D.V. Calvert. 2002. Assessing copper thresholds for phytotoxicity and potential dietary toxicity in selected vegetable crops. J. Environ. Sci. Health, Part B 37:625–635. doi:10.1081/PFC-120015443 Young, I.M., and J.W. Crawford. 2004. Interactions and self-organization in the soil–microbe complex. Science 304:1634–1637. doi:10.1126/science.1097394
