ORIGINAL ARTICLE

Geotechnical systems that evolve with ecological processes

Jason DeJong Mark Tibbett Andy Fourie

Received: 11 September 2013 / Accepted: 16 June 2014

Springer-Verlag Berlin Heidelberg 2014

Abstract Geotechnical systems, such as landfills, mine

tailings storage facilities (TSFs), slopes, and levees, are

required to perform safely throughout their service life,

which can span from decades for levees to in perpetuity

for TSFs. The conventional design practice by geotechnical

engineers for these systems utilizes the as-built material

properties to predict its performance throughout the

required service life. The implicit assumption in this design

methodology is that the soil properties are stable through

time. This is counter to long-term field observations of

these systems, particularly where ecological processes such

as plant, animal, biological, and geochemical activity are

present. Plant roots can densify soil and/or increase

hydraulic conductivity, burrowing animals can increase

seepage, biological activity can strengthen soil, geochem-

ical processes can increase stiffness, etc. The engineering

soil properties naturally change as a stable ecological

system is gradually established following initial construc-

tion, and these changes alter system performance. This

paper presents an integrated perspective and new approach

to this issue, considering ecological, geotechnical, and

mining demands and constraints. A series of data sets and

case histories are utilized to examine these issues and to

propose a more integrated design approach, and consider-

ation is given to future opportunities to manage engineered

landscapes as ecological systems. We conclude that soil

scientists and restoration ecologists must be engaged in

initial project design and geotechnical engineers must be

active in long-term management during the facilitys ser-

vice life. For near-surface geotechnical structures in par-

ticular, this requires an interdisciplinary perspective and

the embracing of soil as a living ecological system rather

than an inert construction material.

Keywords Geotechnical engineering Soil science Ecological engineering Ecological restoration Adaptivemanagement Terrestrial ecology Mining engineering Biological systems Mine tailings Landfills Slopes Levees Dams Bioturbation Ecosystem engineers

Introduction

Geotechnical systems, such as landfills, tailing storage

facilities (TSFs), slopes, and levees, are required to per-

form safely throughout their service life, which can span

from decades for levees to in perpetuity for mine tailing

storage facilities and waste rock dumps. The conventional

geotechnical design practice, and indeed the engineers

outlook itself, for these systems historically assumes that

properties do not generally change through time, and if

they do (by consolidation for example), they do so pre-

dictably. As a result, they typically utilize the as-built, or

in situ, material properties to predict performance

throughout the required service life. Once constructed,

these systems are rarely re-characterized during their ser-

vice life unless either the loading or performance criteria

are changed. The level of performance monitoring also

J. DeJong (&)Civil and Environmental Engineering,

University of California, Davis, USA

e-mail: jdejong@ucdavis.edu

M. Tibbett

Department of Environmental Science and Technology,

Cranfield University, Cranfield MK43 0SZ,

Bedfordshire, England

A. Fourie

School of Civil, Environmental and Mining Engineering,

The University of Western Australia, Crawley, Australia

123

Environ Earth Sci

DOI 10.1007/s12665-014-3460-x

varies, generally proportionally to the level of risk and cost

of failure. Implicit in the practical implementation of this

engineering approach is the assumption that the soil

properties will not change (or will change predictably due

to some mechanical process) from the as-built conditions

and that the system loading realized will match that esti-

mated in design. Ecological systems, in particular the res-

toration of terrestrial ecological systems, fundamentally

expect and depend on soil changing through time. Any

mass of soil, no matter how processed and constructed, will

change with time as part of natural pedogenic processes.

Based on the original quantitative work of Jenny (1941)

describing soil formation as a function of climate, biota,

topography, parent material and time, we can presume all

soil-like material used in near-surface geotechnical engi-

neering schemes will be subject to change: it is just a

matter of time.

For the terrestrial ecologist and soil scientist, the ques-

tion is not Will the soil change through time?, but rather

How will the soil change through time?, What end-

point or climax ecosystem will the site eventually arrive

at?, How can the site be initiated and managed in such a

way that the target stable ecosystem is arrived at sooner?.

It is known and appreciated that plant, animal, microbial,

and geochemical activity are present. Through time, these

processes can induce more change in soil properties than

do construction practices. This is due to the soil matrix

becoming a habitat for organisms and to the enormous

biological diversity in form and function that soil can

support (Bardgett and Wardle 2010; Lavelle and Spain

2001). The activity of these organisms plays a significant

role in the initial terrestrial ecosystem development

through the process of bioturbation, the importance of

which has become increasingly recognized in recent years

(Schaaf et al. 2011; Wilkinson et al. 2009; Meysman et al.

2006). In parallel with these processes, abiotic processes

such as rapid weathering of aggregates in a cover system

can also result in rapid changes to the physical character-

istics of the material. These weathered products can hinder

and laterally deflect water movement in the covers,

potentially resulting in localized zones of very high mois-

ture content.

These contrasting perspectives occur, and interact,

within the soil body simultaneously. Geotechnical con-

struction processes largely dictate initial post-construction

conditions while ecological processes dictate change of the

system throughout its service life. However, in practice

these respective roles are not often considered in an inte-

grated manner. At best, they are considered piece-wise

sequentially, with geotechnical engineering testing and

construction occurring first, followed by some modest level

of ecological monitoring and management. Given the

performance and societal demands for sustainable solu-

tions, these processes need to be integrated in a more

holistic manner.

One example of where geotechnical and ecological

processes are both being considered, studied, and, to some

extent, integrated, is the bauxite mining and land man-

agement process currently implemented in Australia. As

illustrated in Fig. 1, the surficial strip-mining process

effectively removes the top-soil and sub-soil in order to

access the ore body. This process is known as soil double-

stripping (Bell 2001; Koch 2007; Tibbett 2010) and rec-

ognizes the biological significance of the upper soil

Fig. 1 Bauxite mining material(soil, ore, residue) handling

shown in a direct-return model.

Ideally the topsoil and subsoil is

placed directly onto the newly

prepared landscape without

stockpiling; this is the most

effective manner of retaining

soil fertility. The ore is typically

moved by conveyor belt to the

refinery where Bayer

extraction of alumina leaves a

caustic residue disposed of in

containment facilities.

Integration of geotechnical and

ecological processes are

typically well considered in the

mined landscape but less so in

the bauxite residue disposal

facilities where geotechnical

consideration dominate

Environ Earth Sci

123

horizons. These soil layers are not stockpiled but laid on a

previously mined area prepared for restoration to begin.

This system is commonly referred to as a direct-return soil

handling model, and it is the most effective manner to

retain soil fertility (Ward 2000; Tibbett 2010). Once

placed, measures including deep ripping, surface grading,

habitat creation, seeding, planting, and fertilization are

implemented to facilitate a return to something closely

resembling the pre-mining native ecosystem within a

couple of decades (Koch 2007; Koch and Hobbs 2007).

After the ore is removed the site no longer has a specific

geotechnical function (aside from site and slope stability),

and therefore focus is placed on the ecological restoration

of the site. The second part of the post-mining process

consists of transport of the ore body to a processing plant,

extraction of the valued mineral, and then disposal of the

bauxite residue. The bauxite residue is a difficult material

to handle and utilize, with low strength, high water content,

and high pH (Snars and Gilkes 2009). It is typically placed

in large containment facilities indefinitely, with no known

disposal solution. At these locations geotechnical stability

of the residue disposal areas is of primary importance. The

ecological redevelopment at these sites has been given a

lower priority, and these sites are often out of view to the

public, unless they fail completely when they grab the

headlines (e.g. bauxite residue containment facility failure

in Kolontar, Hungary, October 2010).

The above bauxite mining example, while representing a

more progressive approach than in many other fields, still

reflects an unbalanced approach common in geotechnical

and mining systems when considering residue containment.

In general, the engineering development is the focus, and

where and when necessary and appropriate, the ecological

aspects are considered secondary.

This paper examines the apparently conflicting per-

spectives and incompatible roles of geotechnical engi-

neering and terrestrial ecological restoration and

management to increase awareness of the level of inter-

action between these processes that in fact occur in the

field, and to consider how aspects within the ecological

discipline can be better integrated within geotechnical

engineering to obtain more reliable and sustainable solu-

tions for society. This occurs by first considering the

ubiquitous importance of soil, which is a finite resource,

and societys increasing demands for more sustainable

engineering solutions. The traditional perspectives held by

terrestrial ecologists (e.g. soil scientists and plant and

animal biologists) and geotechnical engineers are then

presented. An overview of the conventional design meth-

odology in geotechnical engineering practice is followed

by a series of system examples where soil properties and

parameters change throughout time. The fundamental

ecosystem parameters that drive ecological processes and

can be controlled are identified. This culminates in iden-

tifying how the different alternatives for management

during ecological restoration may be integrated in a more

holistic manner.

Background

Societal expectation for sustainable solutions

Sustainability concepts are gaining widespread acceptance

as they embody societys increased awareness of the fra-

gility of earth and the impact of humans activities on it.

High energy practices are being reduced and regulated as it

is increasingly recognized that resources are finite and the

impact and costs of these activities are compounding (e.g.

IEA 2009; IPCC 2007). In many first world countries new

construction is evaluated against carbon footprint, energy

efficiency, and total life cycle analysis criteria (e.g. CE-

EQUAL 2008; LEED 2009). The historical project

assessment based solely on financial investment is incom-

plete. Projects must now be evaluated against the social

obligation; we have to limit environmental impact, a

holistic assessment of energy embodied in the project, the

projects carbon and water footprint, the probabilities of

failure and rebuilding throughout the project service life,

and the ability for material and site reuse for future gen-

erations (BERR 2008; ICE 2009).

Soil itself, which historically has been considered a

ubiquitous infinite resource, is now being recognized as a

finite resource. Measures at national and international

levels are being adopted to ensure that the management and

use of soil is considered equally with other resources

(Commission of the European Communities 2006; Scottish

Soil Framework 2009). Assessment of the quality and

health of soil at a given project site is assessed prior to

construction, and an assessment is made regarding the

expected condition and quality of the soil once the project

on site is terminated and the site prepared for reuse.

Exactly how preservation of soil health at a site is

accomplished is not fully known (Kibblewhite et al. 2008;

Pankhurst et al. 1997), but current engineering practices

wherein ecological aspects are largely ignored are

insufficient.

An additional issue compounding assessment of a sites

sustainability is the service life over which performance

is expected. The expected service life for cut slopes for

roadways, embankments for overpasses, foundation sys-

tems, etc. is typically about 5075 years. Landfill regula-

tions require safe reliable containment over at least

100 years (Koerner 2005). For levees and dams the

expected life is on the order of 50200 years. Perhaps most

extreme are mining sites, where in some cases containment

Environ Earth Sci

123

performance must last in perpetuity (indefinitely) (Fou-

rie and Mibbett 2007). These time frames are beyond a

politicians term in office, a given engineers or regulators

career, a management policys regulatory authority, and

possibly a companys lifespan. What time span then,

should be used to determine whether a solution/design is

sustainable, and who is financially responsible for manag-

ing the site to ensure the targeted level of sustainability is

achieved?

Differing perspectives: terrestrial ecology vs.

geotechnical engineering

The perspectives on soil between terrestrial ecologists and

geotechnical engineers begin to differ at fundamental lev-

els and length scales. Terrestrial ecologists view soil as a

complex, living system, with many components (Fig. 2a).

Any materials newly placed (or moved) near the land

surface are expected to change through time through the

process of pedogenesis (Jenny 1980) (although consider-

ations of timeframes may vary depending on the nature of

materials concerned). Roots grow, water is absorbed, par-

ticles are displaced, microbes decompose, gas is released,

etc. In contrast the geotechnical engineer concentrates

primarily on the inorganic minerals within soil and the

extent of water present (Fig. 2b). These silt, clay, and sand

particles, and their assemblages, are assumed to remain

largely unchanged through time and to be the primary

source driving the measured engineering properties

(Mitchell and Soga 2005).

The above perspectives are admittedly simplistic and

undoubtedly individuals within each respective field have

broader views. Some terrestrial ecologists would explicitly

account for particle mineralogy as it may affect nutrients

available for plant growth. For example, soils dominated

by kaolinitic clays will have a lower innate fertility than

soils dominated by smectitic clays (Ashman and Puri 2002;

Cresser et al. 1993). Similarly, some geotechnical engi-

neers are beginning to consider the role of biological

activity (e.g. Gray and Sotir 1996; Rowe 2005; Mitchell

and Santamarina 2005; DeJong et al. 2006, 2010; van Pa-

assen et al. 2010) and there is an allocation in soil classi-

fication systems for highly organic peat soils (e.g. USCS

Classification Method, American Society for Testing and

Materials 1985). Nonetheless, the above perspectives do

reflect the common perspectives within each discipline and

Fig. 2 a View of soil from anecological and soil science

perspective and b basiccomponents of soil from a

geotechnical perspective with

i clay platelets, ii sand grains,and iii a mixed soil (figuresfrom Holtz et al. 2010)

Environ Earth Sci

123

the manner in which each is traditionally taught in under-

graduate and (post-) graduate courses.

The differing viewpoints between terrestrial ecology and

geotechnical engineering are further evident when com-

paring the properties and functions of soil that each disci-

pline views as important and measures. Examples (but

certainly not an exhaustive list) of these different properties

are shown in Table 1. The center column contains terms

that may be considered universal, or common, across many

disciplines. Discipline specific terminology and the pur-

pose/function of each property/parameter for terrestrial

ecology and geotechnical engineering are presented to the

left and right, respectively. For example, permeability (the

rate at which water can flow through soil) may be referred to

as infiltration rate at the soil surface (reflecting ability of

water to penetrate soil downward from precipitation), and is

important for estimating soil water recharge versus over-

land-flow after rainfall events. Here, for example, water

infiltration might be measured to estimate plant available

water content (Petersen and Stringham 2008; Rasoulzadeh

and Yaghoubi 2010). The geotechnical engineer may

instead measure the hydraulic conductivity in order to

determine the rate of groundwater (or contaminant) flow or

to estimate the rate at which surface settlement due to

consolidation of clay will occur (Holtz et al. 2010).

The differing views of soil and the importance of given

properties reflect the objective of each discipline. Simply,

terrestrial ecologists examine soil to better assess the extent

to which it can sustain life (a habitat for organisms),

whereas geotechnical engineers test it to determine suitable

properties as an engineering material. Integration of these

two differing perspectives requires a brief review of the

state-of-practice for geotechnical design.

Geotechnical design practice

Geotechnical practice traditionally employs a deterministic

design methodology wherein the most probable loading

conditions and site characteristics are utilized. Site charac-

teristics, namely the stratigraphic profile with depth and the

soil properties of each respective geologic unit, are deter-

mined through limited field and laboratory characterization.

Representative, average values are used for design. The level

of uncertainty in characterization of soil is much larger than

Table 1 Examples of commonproperties of soils with

contrasting terms, functions or

measurements according to

discipline

a Discipline specific

terminology in addition to

common terminology

Terrestrial ecology (soil science) Technical/

common

terminology

Geotechnical engineering

Purpose/function Discipline

terminologyaDiscipline

terminologyaPurpose/function

Relative proportions soil

particles/separates

Texture Particle size Average size,

grain size

distribution

1st indicator of

engineering

properties

Indicator of bulk soil

physical behavior

Arrangement

of pores and

peds

Matrix/

structure

Void ratio,

fabric

1st indicator of soil

stability

Indicator of reactivity,

weathering and physical

behavior

Clay

mineralogy

Mineralogy Mineral

composition

1st indicator of

stiffness

Describes stability of peds

under wetting and

movement

Plastic limit/

liquid limit/

water stable

aggregates

Material

indices

Index properties

(Atterberg

limits)

Describes effect

presence of water

has on engineering

properties

Indictor of root penetration

and gas diffusivity; used

for concentration to mass

calculations

Dry bulk

density

Density Dry density Compatibility,

stiffness, strength

How gases, liquids, or plant

roots penetrate soil;

important for gas and

solute diffusion

Surface

infiltration

rate,

hydraulic

conductivity

Permeability Hydraulic

conductivity

Groundwater/

contaminant flow,

rate of

consolidation

Loss of porosity/voids;

effects gas/solute diffusion

and root penetration

Compaction Compressibility Small strain

stiffness

Settlement,

magnitude of

consolidation

Estimates root penetration

and compaction

Penetration

resistance

Strength Drained or

undrained,

monotonic or

cyclic

strengths

Basis for stability

and capacity

design of most

systems

Environ Earth Sci

123

that for materials used in engineering that are manufactured

in controlled conditions to specific property specifications

(e.g. steel, plastics, concrete) (Baecher and Christian 2003).

Similarly, the predicted loading conditions are inherently

variable as they are driven by natural processes (e.g. wind,

earthquake loading) and variable human behavior (e.g.

traffic levels). To accommodate uncertainty, a safety factor,

wherein the predicted load is, for example, doubled during

design is traditionally implemented (Terzaghi et al. 1996).

In recent years a more advanced probabilistic perfor-

mance based design methodology has developed to more

rigorously account for the uncertainty inherent in both the

site characteristics (i.e. soil properties, stratigraphic vari-

ability across site) and the loading conditions (Baecher and

Christian 2003). This mathematically rigorous framework

enables the level of uncertainty to be applied in a theo-

retical manner to specific design inputs, allowing para-

metric analysis of how the uncertainty in a given variable

affects system performance. This approach then also pro-

vides quantitative assessment of the level of performance

that may be expected, the probability of failure occurring,

and the cost associated with failure. Examples are pre-

sented in Fig. 3. Two geotechnical structures that will be

discussed further herein and that are indicated in Fig. 3 are

foundation systems, and dams and levees. Expectedly, the

probability of failure decreases as the cost, both financially

and in terms of human lives, increases. There appears to be

a frequency threshold of societys tolerance for occurrence,

and this tolerance decreases with increasing cost. Evi-

dently, every system will eventually fail.

In both above design methodologies only the current

engineering soil properties are typically used. The proba-

bilistic method does account for variability in properties,

but usually this variability captures the spatial component

rather than temporal changes. This is due in part to tem-

poral changes in soil properties due to ecological processes

being relatively unknown and difficult to quantify relative

to other engineering materials (e.g. steel where cyclic

fatigue loading can be reliably quantified), and the ability

to repair/replace damaged materials is relatively easy.

Though difficult, it is necessary to begin to incorporate

temporal changes in soil properties within the design

methodologies since clear evidence demonstrates how the

performance of geotechnical systems and soil properties

vary through time due to ecological processes.

Examples of system change during service life

There are numerous examples where failure occurred due

to (adverse) changes in soil properties with time. The

examples below demonstrate how ecological processes that

evolve toward re-establishment of a stable, natural eco-

logical system following construction can significantly

change soil properties and compromise long-term

performance.

Water percolation increase in Rum Jungle rock dumps

Three waste rock dumps were covered in 198485 at the

Rum Jungle Uranium mine, which is in the Northern Ter-

ritory of Australia. These earthen covers were designed to

consist of three distinct layers. The layer immediately

above the retained waste rock was a compacted clay layer

(150225 mm thick), the middle layer was a moisture

retention layer, constructed using sandy clay loam

(150250 mm thick) and the surface layer was intended as

an erosion resistant and pore breaking layer that was

150 mm thick and constructed using gravely sand.

The Rum Jungle cover was one of the most advanced

engineered covers in the world at the time. In retrospect the

layers were too thin. However, at the time of construction

(now some 25? years ago), the norm for covering waste

rock deposits was generally to use a thin veneer (of the

order of 200 mm) of whatever soil was available. The

intermediate layer was intended to act as a store and

release layer; this concept is now widely used interna-

tionally for the design of covers of landfills, TSFs and

waste rock dumps.

One of the key performance criteria for the cover was

that it should limit percolation into the covered waste rock

to no more than 5 % of the annual precipitation value. At

one of the waste rock dump covers, a total of nine pairs ofFig. 3 Chart showing risks associated with annual probability offailure for select civil/mining systems (modified from Baecher 1982)

Environ Earth Sci

123

lysimeters were installed for measuring the percolation rate

and these lysimeters were monitored for 18 years after

construction of the cover. Figure 4 shows the average

variation of the percolation rate (expressed as a percentage

of annual rainfall) and the annual rainfall versus time for

these nine pairs of lysimeters.

For the first 9 years the covers performed as designed,

with percolation rates being less than 5 % of the annual

rainfall. However, after this period the percolation rates

increased significantly and were typically between about 8

and 10 % of the annual rainfall. Field observations also

indicated vegetation dieback in some areas of the cover and

it was therefore decided to embark on a detailed field

investigation. The results of this investigation are contained

in the very detailed report by Taylor et al. (2003) and only

a few salient points are discussed here.

The field investigation was performed 18 years after

cover construction and included in situ hydraulic conduc-

tivity tests using a falling head procedure. For all three

layers the finding was that the hydraulic conductivity had

increased by one to three orders of magnitude. Trenches

were excavated and visual inspection showed extensive

galleries from termites and ants, development of roots

throughout the layers, and the development of polygonal

blocks within the compacted clay layer as a consequence of

desiccation drying with many roots extending along the

cracks between these blocks. Subsequent chemical tests

also showed extensive acidification of the cover soils as a

result of capillary action drawing moisture upwards from

the sulphidic waste material.

Hydraulic conductivity increase in alternative covers

assessment program (ACAP) for landfills

The US Environmental Protection Agency (EPA) initiated

the Alternative Covers Assessment Program (ACAP) in

1998 to provide an improved understanding of the hydro-

logical behavior of both conventional covers (those that

include a compacted clay layer or a low permeability

geosynthetic) and alternative covers (typically those based

on the store and release concept or that include an engi-

neered capillary break layer, or indeed both of these) as

final landfill covers. Large-scale lysimeters were con-

structed at fourteen sites across the U.S., with a key

objective being to study performance in a range of cli-

mates, varying from arid to humid and from hot to cold.

Trial covers were constructed at all sites, with side-by-side

comparisons being carried out at most sites. The field trials

were extensively monitored and a key measure was the

percolation rate through the covers. The results have been

extensively reported in the literature and a succinct review

is provided by Albright et al. (2004).

Of particular relevance herein are the results obtained

from field investigations carried out approximately 5 years

after construction of the field trials (Benson et al. 2007).

During construction, samples of the cover soils were

recovered by taking block samples to produce truly

undisturbed samples. The procedure was repeated in

20022004, with most samples being recovered from the

near surface (upper 30 cm), where most changes in prop-

erties were expected. Laboratory tests were performed on

the undisturbed specimens, including saturated hydraulic

conductivity tests and soilwater characteristic curve

(SWCC) determination using a combination of pressure

plate extractors and chilled mirror hygrometers.

The results showed a surprising change from initial, as-

placed conditions. With one or two minor exceptions, the

hydraulic conductivity increased over time, by as much as

10,000 times in one case. It was clear that the lower the

initial, as-placed saturated hydraulic conductivity (ksat), the

greater was the increase in this parameter over time. The

majority of the soils that had initial ksat values of 10-7 cm/s

increased to values between 10-5 and 10-4 cm/s. Covers

designed as barrier systems were no longer behaving as

barriers. Virtually all specimens tested approached ksatvalues of between 10-5 and 10-3 cm/s irrespective of the

soil texture or the prevailing climatic condition (Fig. 5).

Within the parameters varied, a long-term ksat value less

than 10-5 cm/s appears overoptimistic and potentially

unconservative over the course of the system service life.

The measured water retention characteristics of soils at

all test sites also changed significantly. Two key parame-

ters in characterizing the SWCC are the a and n parame-ters, which are inversely related to the air entry suction and

the slope of the SWCC curve, respectively (Leong and

Rahardjo 1997). The a value increased up to two orders ofmagnitude. This indicates a significant decrease in the air

entry value, which corresponds to the formation of larger

pores (Hillel 1998). The saturated volumetric water content

Fig. 4 Variation of mean infiltration rate (solid symbols) for lysime-ters installed in the Rum Jungle cover system and corresponding

annual rainfall (open symbols) versus years after installation

Environ Earth Sci

123

(equivalent to the porosity) also showed an almost uni-

versal increase, confirming the development of larger

pores. There was also a significant decrease in the

parameter n, meaning that the slope of the SWCC became

shallower. This reflects a broadening of the pore size dis-

tribution, which is consistent with the development of

larger pores as evidenced by the increase in a.

Hydraulic conductivity increase at Rio Tinto Alcan

Gove bauxite mine

Rio Tinto Alcan Gove is a bauxite mine that has been in

operation since the 1970s. The surface strip-mining oper-

ation here uses the double-stripping, direct-return soil

handling processes described earlier (Fig. 1) that retains

optimal biological activity in the soil after severe distur-

bance. This type of mining leaves a patchwork of contig-

uous mined pits of different age that allow measurements

to be made across a time-series of sites of different

restoration age (a chronosequence by space-for-time sub-

stitution). Researchers measured the infiltration rate (static

head disc permeameter) over sites from 1 year old to

26 years old (Fig. 6, Spain et al. 2006) as part of a large

study on development of restored native forest ecosystems

at this mine (Spain et al. 2009). The infiltration rate is

synonymous to hydraulic conductivity or permeability

(Table 1). Remarkably, and perhaps quite disturbingly, the

effective permeability at the soil surface increased by five

times in quarter of a century of ecosystem development.

While for this site, such an observation may be seen as a

generally positive development from an ecological per-

spective, had this soil been the cover of a vegetated landfill

site, waste rock dump or tailing storage facility it might be

a worrying development. In such a case, a fundamental

geotechnical measurement made for the design of an

impoundment would be quite different 25 years after

commissioning.

Soil structure/fabric stability at Oaky Creek Coal mine

A simple but instructive case shows the speed with which

soil can change. The measurements concerned occurred as

part of an ecological assessment of internally draining re-

vegetated landforms at Oaky Creek Coal mine (Tongway

and Hindley 1998). The test performed was a simple

slaking test that assesses soil structural stability under

immersion in water (modified after Emerson 1967). Soils

of restoration ages between 1 and 8 years were scored

according to their stability or propensity for slaking, with a

score of 0 indicating complete disaggregation and non-

coherence, and 4 indicating complete stability (Table 2).

The results showed that in a period of only 8 years the soils

that were fundamentally unstable became at least three

times more stable. Notably, the run-on areas, where sedi-

ment, carbon and moisture tend to accumulate (all stimu-

lating microbial activity), the soil stability index reached a

score of 3.7 (on a scale to 4.0). This is clear evidence that

soils can change in their physical characteristics in rapid

sub-decadal timeframes and in this case in a positive way

both ecologically and geotechnically.

Tree root penetration in Sacramento, California levees

Regulated maintenance of more than 20,000 linear kilo-

meters of levees in California, mostly constructed in a non-

Fig. 5 Post-construction versus as-built saturated hydraulic conduc-tivity for soils after 24 years, as measured in ACAP research

program in the USA (after Benson et al. 2007)

Time (Years)0 5 10 15 20 25 30

Infilt

ratio

n Ra

te (m

m/hr)

0

1000

2000

3000

4000

5000

Fig. 6 Changes in infiltration rate with time in rehabilitated bauxitemine soils (after Spain et al. 2006). Means are based on replicate

observations from each site of a different age class (1, 2, 3, 4, 8, 13,

20, 26 years where n = 6, 9, 8, 6, 7, 6, 6, 6 respectively) R2 = 0.97

Environ Earth Sci

123

engineered manner decades ago, has become a central issue

following multiple levee failures in New Orleans in 2005

(USACE 2007). Failure to meet new requirements can

result in an unacceptable rating, a loss of accredita-

tion leaving cities and people protected by these levees

immediately uninsured and continued national fiscal sup-

port for maintenance tenuous (Harder et al. 2010). Central

to accreditation is management of the naturally evolving

ecological system on the levee while it continues to per-

form its function of retaining water. The particular issue

highlighted herein is the management of tree/shrub growth

on and near the levees. Recent USACE guidelines require

no woody vegetation on the levees or within 15 feet (about

5 m) of the levee toe on either side (USACE 2009).

The motivation behind this regulation is the expectation

that root penetration from nearby vegetation will adversely

affect levee performance, perhaps by root penetration

creating preferential water seepage paths. For new levees,

prevention of vegetation growth is a straightforward pro-

posal. However, removal of well established vegetation is

complicated as the root structure is already established, and

it is unknown whether gradual decay of the root system

from a removed tree would be more harmful.

A study was undertaken to examine the root structure

expanse from an established tree adjacent to a levee. Of

particular interest was how the root system interacts with

alternative levee stabilization measures, in this case a soil

cementbentonite (SCB) slurry cutoff wall (Harder et al.

2010). The study was performed on a levee in Sacramento,

California, where a 2.1 m basal diameter walnut tree was

established 10.6 m from a continuous SCB slurry cutoff

aligned along the center of the levee penetrating 5.1? m

into the levee. During excavation adjacent to the cutoff

wall (on the same side as the tree itself) an established

network of roots that included both live roots as well as

dead roots (damaged from cutoff wall installation in 1991)

was observed. Root sizes ranged from primary 0.63.2 cm

in diameter, with networks of smaller roots distributed on

the face of the cutoff wall (Fig. 7a). From an engineering

perspective a primary concern was whether the roots pen-

etrated through the cutoff wall, thereby compromising its

integrity and performance. Excavation of the opposing face

of the cutoff wall revealed that roots penetrated through the

cutoff wall and were distributed on the opposing face

(Fig. 7b). Further investigation revealed that the vertical

cracks had likely formed due to the annual wetting and

drying cycles near the top of the wall, and that the roots

penetrated the cracks once formed.

This recent full scale field study is one of the first to

clearly demonstrate the extent to which vegetation that is

part of a naturally evolving ecosystem penetrates

throughout an engineering system. The effects of ecosys-

tem development (namely root growth in this case) were

not considered during initial design or during subsequent

remediation when the cutoff wall was installed.

Slope stability and soil shear strength changes with root

penetration

Numerous cases document how clearance of vegetation on

a slope can lead to slope failures which in general are

shallow, translational slides (Bishop and Stevens 1964;

Gray 1970; Rice and Krammes 1970; Waldron 1977;

Watson and OLoughlin 1990; Gray and Sotir 1996). Wu

et al. (1979) reported landslide frequency on Prince of

Wales Island, Alaska, where it was found that slope fail-

ures frequently occurred during periods of heavy autumn

rain a few years after felling of trees. They attributed this

lag period between felling of trees and the onset of land-

sliding to the time required for tree root decay.

These observations have generally led to the conclusion

that the tree roots contributed mechanically (i.e. by rein-

forcement) to stabilization of slopes (Waldron and Da-

kessian 1982), although factors such as rainfall

interception, reduced rates of evaporation, and prevention

of surficial erosion are clearly also important. A number of

previous studies of the reinforcing effect of roots have

utilized direct shear box equipment (Waldron and Dakes-

sian 1982; Wu et al. 1979; Operstein and Frydman 2000).

Fourie (2007) describes results from a series of triaxial

tests on root-reinforced soils, which quantified the

mechanical reinforcing effect of roots. In these tests the

root content was equivalent to only 0.5 kg/m3, which is

much less than typical values of around 23 kg/m3 (Jack-

son et al. 1996) for temperate grasslands. Despite the rel-

atively low root density, the minimum increase in shear

strength was 6 kPa (from a reference value of 15 kPa with

no roots present) and was often greater than this value. As

is well known in geotechnical engineering practice, even

relatively small increases in shear strength can be very

beneficial in preventing the development and propagation

of shallow slope failures. As root development occurs in

Table 2 Mean slake indicator score for landscape types categorizedby their response to water overland-flow from rainfall

Site age Run-off area Run-on area Neutral area

1 0.8 0.6 0.8

4 0.5 1.6 0.8

6 1.7 2.4 2.7

8 2.4 3.7 2.8

These are either: run-off areas, where water is shed; run-on areas,

where water is ponded or neutral zones between the two previous

categories. Scores range from 0 = complete disaggregation and non-

coherence to 4 = complete stability where n = 4 (after Tongway and

Hindley 1998)

Environ Earth Sci

123

the near-surface region of a soil profile, it can thus provide

a reinforcing effect exactly where it is most beneficial, i.e.

where it can prevent the development of shallow slides.

Summary

In all except the last of the above examples (which are far

from exhaustive) some critical ecological process was

ignored during design and/or inadequately managed during

the systems service life. In each system, one or more

components of the ecological system created a change in

soil properties or site conditions that compromised system

performance, whereas the last example is one in which the

soil shear strength was improved. The cause for the over-

sight in the majority of the examples quoted is under-

standable but unacceptable; the extensive changes that

ecological systems can induce in soil properties is complex

and awareness of potential issues is lacking due to insuf-

ficient interaction between these two disciplines. Overall,

the important lesson to be learned is the unavoidable

impact of natural processes on engineered landscapes.

Whether these impacts are always deleterious is not the

issue; it is rather the pressing need for engineers to rec-

ognize that certain realities need to be faced when

designing near-surface soil systems and that purely engi-

neered solutions may not be stable through time.

Consequences

Examples of soil property and parameter change

through time

The change through time of soil properties can be linked, in

many cases, directly to ecological processes at all length

scales. The time period over which properties can change,

and the magnitude and direction in which they change are

unique, and in some cases cannot be predicted or known

(a)

(b)

Fig. 7 a Root network comingfrom walnut tree toward SCB

cutoff wall, and b rootpenetration through crack in

SCB cutoff wall (from Harder

et al. 2010)

Environ Earth Sci

123

ahead of time. Figure 8 contains a series of charts that

schematically indicate changes at length scales from

micrometers (lm) through to meters (m). The left columnrepresents typical/common terrestrial ecological parame-

ters that would be varied while the right column represents

typical/common geotechnical engineering soil properties

that would be measured (reflecting further the different

perspectives presented in Fig. 2; Table 1). The trends

shown in each figure are based on published studies (ref-

erences provided in Figure legend), but are presented

schematically to emphasize the extent of changes to

parameters/properties across all length scales. As evident in

Fig. 8, the rate of change through time as well as the

overall magnitude of change is not consistent or constant.

The trends shown indicate broadly predictable increases in

each parameter, regardless of scale. The resulting impli-

cation to geotechnical properties is often less clear. It is

noted that some guidance, however, can be provided when

the corresponding value for a surrounding natural ecosys-

tem reference site can be given (provided the ecological

restoration project is targeting re-establishment of the

natural ecosystem, see Tibbett 2010).

Terrestrial ecological processes drive landscape change

Accommodation of how soil parameters and properties will

vary through time during a projects service life requires an

ability to predict this change to a reasonable extent. It is

unrealistic to expect that the exact change in soil properties

could be predicted (e.g. increase in permeability of 3.5

times). However, awareness of the underlying processes

driving re-establishment of a terrestrial ecosystem can

enable an intelligent estimate of the direction and

approximate magnitude of change that could occur. This

prediction could then be integrated into geotechnical

design to begin to account for changes in soil properties

through time due to ecological restoration.

The primary factor that drives the changes in engineered

landscapes is the flow of carbon into the evolving soil.

Biological carbon fixation (photosynthesis) is the primary

Fig. 8 Changes in properties ofecological (left panels) and

geotechnical (right panels)

parameters across time scales

(assuming no biological

toxicities). Temporal trends

shown by solid lines are based

on observational data from the

literature; trends shown by

dashed lines are based on either

contrasting data sets from

literature or where data is

limited and dotted lines show

anticipated range in trends. The

ecological parameters show

broadly predictable increases in

each measure, regardless of

scale, any of which may affect

geotechnical properties of soils.

Enzyme activities after Spain

et al. (2006) for Chitinase [EC

3.2.1.14]; Spain et al. (2009) for

b -glucosidase [EC 3.2.1.21]and acid phosphatase [EC

3.1.3.41]). Soil microbial

biomass after Spain et al.

(2006); Banning et al. (2008).

Invertebrates after Majer et al.

(1984) and Spain et al. (2010).

Root density after Spain and

Tibbett (2011) and Spain et al.

(2009, 2014). Basel tree area

after Tongway and Ludwig

(2011). Density and

permeability after Benson et al.

(2007). Strength after Operstein

and Frydman (2000), and slope

stability after Gray (1970)

Environ Earth Sci

123

means by which carbon is delivered into the inert surficial

substrate as it develops into an incipient soil. This occurs

via roots, their symbionts and leaf litterfall. Carbon is the

primary source of energy in ecosystems and its presence

drives a range of processes that will change the physical

and chemical properties of the substrate into which it is

introduced (see Fig. 8). The mechanisms of organic carbon

deposition into a substrate range from simple root pene-

tration, which include the carbon flows into symbiotic

fungal partners of roots (mycorrhizasSmith and Read

2008), to the leakage of carbon from roots (Bottner et al.

1999; Jones et al. 2004) and litter deposited as dead roots,

woody debris and leaves (Hutsch et al. 2002; Harmon et al.

1986; George et al. 2010). All of these can contribute to a

physical reorganization and chemical alteration of the

developing soil (see for example Feeney et al. 2008; Spain

et al. 2006). The presence of carbon itself will stimulate the

colonization of a wide range of heterotrophic microor-

ganisms (such as bacteria, fungi, protozoa and nematodes)

and litter feeding and predatory invertebrates (including

collembola, beetles, earthworms ants and termitesMajer

et al. 1984; Spain et al. 2010) and a food web will become

established (Ferris 2010; Rygiewicz et al. 2010). The

establishment of a food web will drive further development

of the soil ecosystem as organisms at various trophic levels

(position in the food web) affect their environment to best

suit themselves. This process has become known as niche

construction (Kylafis and Loreau 2008) and may cause

substantial changes to soils that may be positive or negative

in terms of geotechnical (anthropomorphic) outcomes but

perfectly suited to the needs of the organisms (the agents of

change).

When carbon cycling in a plantsoil system is stable it is

likely that the ecosystem will be close to a dynamic

equilibrium. The process of ecological restoration can then

be considered a gradual process of restoring stable carbon

cycling in the soil system. It is important to recognize how

this contrasts with the geotechnical engineering perspective

of soil. Engineers seek to identify and use soils in con-

struction with a negligible amount of organic content in

order to obtain improved engineering properties (as

reflected in soil classification systems such as ASHTOO

and USCS). In effect, this creates an initial constructed

system that is well removed from the natural ecosystem,

positioning the system to undergo significant change as the

ecosystem is inevitably re-established.

Managing an evolving engineered land system

The extent of change in soil parameters/properties, and

therefore the overall system performance through time, is

not abandoned to nature as it takes its natural course. Most

geotechnical systems are managed at some level

throughout the service life, and management measures can

include ecosystem management (though it is often not

thought of this way in geotechnical engineering). Examples

of this include the removal of all vegetation on large

earthen dams, culling of vegetation with trunk diameters

larger than 300 (about 7.5 cm) on urban levees (USACE2009), mowing of grasses on landfills, and other various

forms of ecological adaptive management (Holling

1978) that might allow the trajectory of the developing

ecosystem to be modified.

Conceptually, a series of alterative management options

exist as shown schematically in Fig. 9a. For reference, the

traditional assumed geotechnical condition is shown to be

unchanging with time. If nature was left to run its course

then a native ecosystem (or target ecosystem if an alter-

native final condition is desired) will be eventually attained

through a process known as ecological succession (Connell

and Slatyer 1977). This typically takes several decades or

even centuries. Proactive management seeks to restore the

native (or target) ecosystem in a shortened time frame.

Such measures often undertaken include managing the soil

by deep ripping and fertilization, manipulating the flora by

promoting key species and repressing others through

(a)

(b)

Fig. 9 The potential change (increase or decrease from initialconditions) of biotic and abiotic parameters (e.g. hydraulic conduc-

tance or aggregate stability) or ecosystem properties (e.g. functional

diversity or plant density) with time as a terrestrial ecosystem

develops. a For different perceptions and management options andb differences that might occur and could be managed for or against

Environ Earth Sci

123

sowing seeds, out planting from nurseries and weeding

(Koch 2007, 2011; Tibbett 2010; Lardner et al. 2011). An

alternative to conventional restoration (targeting a native

ecosystem) is the establishment of a new, or novel, eco-

system. A novel ecosystem is one that is different in its

community of organisms and, importantly, its ecological

functions than other undisturbed systems (Hobbs et al.

2009). This is increasingly recognized as an almost inevi-

table consequence of environmental alteration through land

use change, and allows landscapes to be designed and

managed that serve a predetermined series of geotechnical

and environmental parameters, while retaining intrinsic

ecological value. Again, measures can be taken to establish

this ecosystem in an accelerated manner. Moreover, the

final ecosystem established would ideally meet the project

site requirements (e.g. no deep rooted trees, no burrowing

mammals, vegetation with high ash alkalinity), therefore

requiring less intervention once the ecosystem is estab-

lished. Examples of where these different levels of eco-

system management have been put into practice range from

zero management in the natural successional recovery of

Mount St. Helens (Dale et al. 2005), bauxite mining for

complete ecological restoration (Tibbett 2010), levees

and TSFs for Novel ecosystems with some established

vegetation (UASCE 2009; Rentel and Rental 2009), and

landfills and earthen dams for extreme management such

that the geotechnical design assumption is realized in the

field to the extent possible (Fell et al. 2005).

Though the process of ecological restoration is con-

ceptually clear, the actions taken to accelerate establish-

ment of a natural, living ecological system make

realization of the concept challenging and are perhaps best

couched in the new broader concepts of Intervention

Ecology (Hobbs et al. 2011). Figure 9b schematically

exemplifies the different realization that may occur, with

the Ideal trend representing the objective. Any of these

trends represent some form of intervention ecology where

we employ a mechanism to alter the trajectory and out-

come of the final system by managing its functional and

biotic evolution as a land system.

Toward an integrated approach

Conventional design of geosystems largely follows a linear

process in which geotechnical design occurs first, followed

by facility construction (Fig. 10a). Only after construction

is complete are soil scientists and restoration ecologists

consulted for long-term management. This piece-wise

sequential process is a primary source in the above

described failures. Moving forward, it is proposed that a

non-linear approach wherein the facility is considered as an

integrated, engineered biological system rather than a

(a)

(b)

Fig. 10 Changes in engineeredlandscape development in

stages that currently and might

occur under a conventionaldesign approach and b proposedintegrated approach

Environ Earth Sci

123

geotechnical system must be adopted (Fig. 10b). In this

approach soil scientists and restoration ecologists are

engaged from initial design and management to the end of

the facilitys service life, which may be in perpetuity.

Realization of the approach is likely to be unique for every

facility and will require a management perspective that is

interdisciplinary, fluid, and engaged.

Conclusions

The differing view of soil from the soil science and eco-

logical perspectives relative to the geotechnical engineer-

ing perspective highlights the historical division between

these traditional disciplines. The under-appreciation of the

complexity of soil, particularly from the geotechnical

engineering perspective in recognizing soil as a living

ecosystem, has led to a conventional design approach that

is linear and overly simplistic. A number of case histories

demonstrated how lack of consideration of the temporal

ecological aspects of soil has resulted in deterioration of

geosystems. Clearly, the changes in soil properties tem-

porally are complex and not easily predicted. Insight,

however, can gained by considering the carbon fixation

within the soil and using established ecosystems sur-

rounding a project site as a long-term indicator of what

ecological system may be sustainable. Moving forward,

soil scientists and restoration ecologists must be engaged in

the initial project design team and geotechnical engineers

must be active in long-term management during the facil-

itys service life. For near-surface geotechnical structures

in particular, this requires an interdisciplinary perspective

and the embracing of soil as a living ecological system

rather than an inert construction material.

Acknowledgments Funding provided by the United States NationalScience Foundation (#0727463), Geosyntec Inc., and the UC Dis-

covery Grant Program in support of the research by Jason T. DeJong.

Any opinions, findings and conclusions or recommendations expres-

sed in this material are those of the writer(s) and do not necessarily

reflect the views of the National Science Foundation.

References

Albright W, Benson C, Gee G, Roesler A, Abichou T, Apiwantragoon

P, Lyles B, Rock S (2004) Field water balance of landfill final

covers. J Environ Qual 33:23172332

American Society for Testing and Materials (1985) Classification of

Soils for Engineering Purposes: Annual Book of ASTM

Standards, D 2487-83, pp 395408

Ashman MR, Puri G (2002) Essential soil science. Blackwell science

Ltd., Oxford

Baecher BG (1982) Statistical methods in site characterization. In:

Updating Subsurface Sampling of Soils and Rocks and their In-

Situ Testing, Engineering Foundation, Santa Barbara, pp 463

492

Baecher G, Christian J (2003) Reliability and statistics in geotechnical

engineering. Wiley, New York

Banning NC, Grant CD, Jones DL, Murphy DV (2008) Recovery of

soil organic matter, organic matter turnover and nitrogen cycling

in a post-mining forest rehabilitation chronosequence. Soil Biol

Biochem 40:20212031

Bardgett RD, Wardle DA (2010) Aboveground-belowground link-

ages: biotic interactions, ecosystem processes, and global

change. Oxford series in ecology and evolution. Oxford

University Press, Oxford

Bell LC (2001) Establishment of native ecosystems after mining:

Australian experience across diverse biogeographic zones. Ecol

Eng 17:179186

Benson C, Sawangsuriya A, Trzebiatowski B, Albright W (2007)

Postconstruction changes in the hydraulic properties of water

balance cover soils. ASCE JGGE 133(4):349359

BERR (2008) Strategy for sustainable construction, business, enter-

prise and regulatory reform (BERR), HM Government, p 60

Bishop DM, Stevens ME (1964) Landslides on logged areas in

southeast Alaska. U.S. Forest Service Research Paper NOR-1,

Juneau, Alaska, p 18

Bottner P, Pansu M, Sallih Z (1999) Modelling the effect of active

roots on soil organic matter turnover. Plant Soil 216:1525

CEEQUAL (2008) The civil engineering environmental quality

assessment and awards scheme manual, Version 4, CEEQUAL

Ltd., p 114

Commission of the European Communities (2006) Thematic strategy

for soil protection communication from the commission to the

council, the European parliament, The European economic and

social committee and the committee of the regions,

COM(2006)241 final, p 12

Connell JH, Slatyer RO (1977) Mechanisms of succession in natural

communities and their role in community stability and organi-

zation. Am Nat 111(982):119144

Cresser M, Killham K, Edwards T (1993) Soil chemistry and its

applications. Cambridge University Press, Cambridge

Dale VH, Swanson FJ, Crisafulli CM (eds) (2005) Ecological

responses to the 1980 eruption of Mount St. Helens, Springer

DeJong JT, Fritzges MB, Nusslein K (2006) Microbial induced

cementation to control sand response to undrained shear. ASCE J

Geotech Geoenviron Eng 132(11):13811392

DeJong JT, Mortensen BM, Martinez BC, Nelson DC (2010) Bio-

mediated soil improvement. Ecol Eng 36:197210

Emerson WW (1967) A classification of soil aggregates based on

their coherence in water. Aust J Soil Res 5:4757

Feeney DS, Crawford JW, Daniell T, Hallett PD, Nunan N, Ritz K,

Rivers M, Young IM (2008) Three-dimensional microorganiza-

tion of the soilrootmicrobe system. Microbiol Ecol 52:151158

Fell R, MacGregor P, Stapledon D, Bell G (2005) Geotechnical

engineering of dams. Taylor & Francis, p 912

Ferris H (2010) Contribution of nematodes to the structure and

function of the soil food web. J Nematol 42:6367

Fourie AB (2007) The engineering contribution of vegetation to the

stability of cover systems. In: Fourie AB, Tibbett M, Wiertz J

(eds) Proceedings of 2nd International Seminar on Mine Closure,

October 2007, Santiago, Chile, pp 483493

Fourie AB, Mibbett M (2007) Post-mining landforms: engineering a

biological system. Keynote Lecture. In: Fourie AB, Tibbett M,

Wiertz J (eds) 2nd International Seminar on Mine Closure,

October 2007, Santiago, Chile, pp 312

George SJ, Kelly RN, Greenwood PF, Tibbett M (2010) Soil carbon

and litter development along a reconstructed biodiverse forest

chronosequence of South-Western Australia. Biogeochemistry

101:197209

Gray DH (1970) Effects of forest clear-cutting on the stability of

natural slopes. Bull Assoc Eng Geol 7(12):4566

Environ Earth Sci

123

Gray DH, Sotir RB (1996) Biotechnical and soil bioengineering

stabilization. Wiley, New York, p 378

Harder LF, Kroll R, Claassen V, Buck PEF, Berry AM (2010)

Investigation of tree root penetration into a levee soil-cement-

bentonite slurry cutoff wall. In: Annual Convention of the

Association of State Dam Safety Officials, Seattle

Harmon ME, Franklin JF, Swanson FJ, Sollins P, Gregory SV, Lattin

SD, Anderson NH, Cline SP, Aumen NG, Sedell JG, Lienka-

emper GW, Cromack K Jr, Cummins KW (1986) Ecology of

coarse woody debris in temperate ecosystems. Adv Ecol Res

15:133302

Hillel D (1998) Environmental soil physics. Academic Press, San

Diego

Hobbs RJ, Higgs E, Harris JA (2009) Novel ecosystems: implications

for conservation and restoration. Trends Ecol Evol 24:599605

Hobbs RJ, Hallett LM, Ehrlich PR, Mooney HA (2011) Intervention

ecology: applying ecological science in the twenty-first century.

Bioscience 61:442450

Holling CS (ed) (1978) Adaptive environmental assessment and

management. Wiley, New York

Holtz RD, Kovacs WD, Sheahan TC (2010) An Introduction to

Geotechnical Engineering. Prentice Hall, NJ

Hutsch BW, Augustin J, Merbach W (2002) Plant rhizodeposition: an

important source for carbon turnover in soils. J Plant Nutr Soil

Sci 165:397407

ICE (2009) The state of the nation, low carbon infrastructure.

Institution of civil engineers, p 15

IEA (2009) World energy outlook 2009. International energy agency

IPCC (2007) Fourth assessment report: climate change, intergovern-

mental panel on climate change

Jackson RB, Canadell J, Ehleringer JR, Mooney HA, Sala OE,

Schulze ED (1996) A global analysis of root distributions for

terrestrial biomes. Oecologia 108:389411

Jenny H (1941) Factors of soil formation. A system of quantitative

pedology. McGraw Hill Book Company, New York

Jenny H (1980) The soil resource: origin and behaviour. Springer,

New York, p 377

Jones DL, Hodge A, Kuzyakov Y (2004) Plant and mycorrhizal

regulation of rhizodeposition. New Phytol 163:459480

Kibblewhite MG, Ritz K, Swift MJ (2008) Soil health in agricultural

systems. Philos Trans R Soc Lond B Biol Sci 363:685701

Koch JM (2007) Alcoas mining restoration process in South Western

Australia. Restor Ecol 15(Suppl):S11S16

Koch JM, Hobbs RJ (2007) Synthesis: is Alcoa successfully restoring

a jarrah forest ecosystem after bauxite mining in Western

Australia? Restor Ecol 15(Suppl):137144

Koerner RM (2005) Designing with geosynthetics. Prentice Hall,

New York

Kylafis G, Loreau M (2008) Ecological and evolutionary conse-

quences of niche construction for its agents. Ecol Lett

11:10721081

Lardner TD, Worthington TR, Braimbridge MF, Vlahos S, Tibbett M

(2011) Optimising soil physical properties for rehabilitation of

mined land: effects of tine type on soil strength and root

proliferation. In: Fourie AB, Tibbett M, Beersing A (eds)

Proceedings of the Sixth International Conference Mine Closure,

Australian Centre for Geomechanics, Perth, pp 153164

Lavelle P, Spain AV (2001) Soil ecology. Kluwer Academic,

Dordrecht, p 654

LEED (2009) New construction and major renovations rating system.

U.S. Green Building Council, p 88

Leong EC, Rahardjo H (1997) Permeability functions for unsaturated

soils. J Geotech Geoenviron Eng 123(12):11181126

Majer JD, Day JE, Kabay ED, Perriman WS (1984) Recolonization

by ants in bauxite mines rehabilitated by a number of different

methods. J Appl Ecol 21:355375

Meysman FJR, Middelburg JJ, Heip CHR (2006) Bioturbation: a

fresh look at Darwins last idea. Trends Ecol Evol 21(12):

688695

Mitchell JK, Santamarina JC (2005) Biological considerations in

geotechnical engineering. J Geotech Geoenviron Eng 131(10):

12221233

Mitchell JK, Soga K (2005) Fundamentals of soil behavior. Wiley,

New York

Operstein V, Frydman S (2000) The influence of vegetation on soil

strength. Ground Improv 4:8189

Pankhurst C, Doube MB, Gupta VVSR (eds) (1997) Biological

indicators of soil health. Wallingford, UK

Petersen SL, Stringham TK (2008) Infiltration, runoff, and sediment

yield in response to western juniper encroachment in southeast

Oregon. Rangel Ecol Manag 61:7481

Rasoulzadeh A, Yaghoubi A (2010) A effect of cattle manure on soil

physical properties on a sandy clay loam soil in North-West Iran.

J Food Agric Environ 8:976979

Rentel U, Rental M (2009) Determining the rehabilitation success of

the old tailings storage facility of Navachab Gold Mine, Karibib,

Namibia. In: Fourie AB, Tibbett M (eds) Proceedings of the fifth

international conference mine closure, Perth, Australia,

pp 109121. Australian Centre for Geomechanics, Perth

Rice RM, Krammes JS (1970) Mass-wasting processes in watershed

management. In: Proceedings on symposium interdisciplinary

aspects of watershed management, Bozeman, Montana, August

1970. ASCE, New York, pp 231260

Rowe RK (2005) Long-term performance of contaminant barrier

systems, 45th Rankine Lecture. Geotechnique 55(9):631678

Rygiewicz PT, Monleon VJ, Ingham ER, Martin KJ, Johnson MG

(2010) Soil life in reconstructed ecosystems: initial soil food web

responses after rebuilding a forest soil profile for a climate

change experiment. Appl Soil Ecol 45:2638

Schaaf W, Anton Fischer OB, Gerke HH, Gerwin W, Grunewald U,

Hollander HM, Kogel-Knabner I, Mutz M, Schloter M, Schulin

R, Veste M, Winter S, Huttl RF (2011) Patterns and processes of

initial terrestrial-ecosystem development. J Plant Nutr Soil Sci

174:229239

Scottish Government (2009) The Scottish soil framework, p 64

Smith SE, Read DJ (2008) Mycorrhizal symbiosis, 2nd edn. Elsevier,

London

Snars K, Gilkes RJ (2009) Evaluation of bauxite residues (red muds)

of different origins for environmental applications. Appl Clay

Sci 46:1320

Spain AV, Tibbett M (2011) Substrate conditions, root and arbuscular

mycorrhizal colonisation of landforms rehabilitated after coal

mining, sub-tropical Queensland. In: Fourie AB, Tibbett M,

Beersing A (eds) Proceedings of the sixth international confer-

ence mine closure, Perth, Australia. Australian Centre for

Geomechanics, Perth, pp 199208

Spain AV, Tibbett M, Hinz DA, Ludwig JA, Tongway DJ (2014) The

Mining-restoration system and ecosystem development follow-

ing bauxite mining in a biodiverse environment in the seasonally

dry tropics of Northern Australia. In: Tibbett M (ed) Mining in

Ecologically Sensitive Landscapes, CRC Press (in press)

Spain AV, Hinz DA, Ludwig J, Tibbett M, Tongway D (2006) Mine

closure and ecosystem development: Alcan Gove bauxite mine,

NT, Australia. In: Fourie AB, Tibbett M (eds) Proceedings of the

first international seminar on mine closure, pp 299308. Aus-

tralian Centre for Geomechanics, Perth

Spain AV, Ludwig J, Tibbett M, Tongway D (2009) Ecological and

minesoil development studies at the Rio Tinto Alcan Gove Mine

site, Northern Territory. Centre for Land Rehabilitation, Perth

Spain AV, Hinz D, Tibbett M (2010) Colonisation of rehabilitated

lands by termites (Dictyoptera), RTA Gove bauxite mine, NT,

Australia. In: Fourie AB, Tibbett M, Wiertz J (eds) Proceedings

Environ Earth Sci

123

of the fifth International conference mine closure, Perth, Austra-

lia, pp 437448. Australian Centre for Geomechanics, Perth

Taylor G, Spain A, Nefiodovas A, Timms G, Kuznetsov V, Bennett J

(2003) Determination of the reasons for deterioration of the rum

jungle waste rock cover, Australian centre for mining environ-

mental research (Brisbane)

Terzaghi K, Peck RB, Mesri G (1996) Soil mechanics in engineering

practice. Wiley, New York

Tibbett M (2010) Large-scale mine site restoration of australian

eucalypt forests after bauxite mining: soil management and

ecosystem development. In: Batty LC, Hallberg K (eds) Ecology

of industrial pollution. Cambridge University Press, UK,

pp 309326

Tongway DJ, Hindley NL (1998) An ecological assessment of

internally draining landforms at Oaky Creek Coal mine. CSIRO

wildlife and ecology report, p 58

Tongway DJ, Ludwig JA (2011) Restoring disturbed landscapes.

Island Press, Washington, DC

United States Army Corps of Engineers (2007) Treatment of

vegetation within local flood-damage-reduction systems. Draft

Final White Paper, p 34

United States Army Corps of Engineers (2009) Guidelines for

landscape planting and vegetation management at levees,

floodwalls, embankment dams, and appurtenant structures.

ETL 1110-2-571, p 32

van Paassen LA, Ghose R, van der Linden TJM, van der Star WRL,

van Loosdrecht MCM (2010) Quantifying biomediated ground

improvement by ureolysis: large-scale biogrout experiment.

ASCE J Geotech Geoenviron Eng 136(12):17211728

Waldron LJ (1977) The shear resistance of root-permeated homoge-

neous and stratified soil. Soil Sci Soc Am J 41:843849

Waldron LJ, Dakessian S (1982) Effect of grass, legume, and tree

roots on soil shearing resistance. Soil Sci Soc Am J 46:894899

Ward SC (2000) Soil development on rehabilitated bauxite mines in

south-west Australia. Aust J Soil Res 38:453464

Watson A, OLoughlin CL (1990) Structural root morphology and

biomass of three age-classes of Pinus radiata. N. Z. J For Sci

20(1):97110

Wilkinson MT, Richards PJ, Humphreys GS (2009) Breaking ground:

pedological, geological, and ecological implications of soil

bioturbation. Earth Sci Rev 97:257272

Wu TH, McKinnell WP III, Swanston DN (1979) Strength of tree

roots and landslides on Prince of Wales Island, Alaska. Can

Geotech J 16:1933

Environ Earth Sci

123

Geotechnical systems that evolve with ecological processesAbstractIntroductionBackgroundSocietal expectation for sustainable solutionsDiffering perspectives: terrestrial ecology vs. geotechnical engineeringGeotechnical design practice

Examples of system change during service lifeWater percolation increase in Rum Jungle rock dumpsHydraulic conductivity increase in alternative covers assessment program (ACAP) for landfillsHydraulic conductivity increase at Rio Tinto Alcan Gove bauxite mineSoil structure/fabric stability at Oaky Creek Coal mineTree root penetration in Sacramento, California leveesSlope stability and soil shear strength changes with root penetrationSummary

ConsequencesExamples of soil property and parameter change through timeTerrestrial ecological processes drive landscape changeManaging an evolving engineered land systemToward an integrated approach

ConclusionsAcknowledgmentsReferences