Contents lists available at ScienceDirect

International Biodeterioration & Biodegradation

journal homepage: www.elsevier.com/locate/ibiod

A review and case studies of factors affecting the stability of woodenfoundation piles in urban environments exposed to construction work

Johanna Elam∗, Charlotte BjördalDepartment of Marine Sciences, University of Gothenburg, Carl Skottsbergs Gata 22 B, 41319, Göteborg, Sweden

A R T I C L E I N F O

Keywords:Bacterial degradationFoundation pilesGround workHistoric buildingsUrban hydrogeologyWood

A B S T R A C T

Wooden foundation piles have been used since Roman times to solidify infirm soils prior to construction of heavymonuments like churches, buildings and bridges. These are known to have a long service life in waterloggedanaerobic environments where most biological wood decay processes, except for slow bacterial degradation, arerepressed. However, large scale urban ground work may change the biogeochemical stability of the soil, in-creasing the microbial decay of the piles and resulting in settling damages to historic buildings. This paperevaluates and synthesizes present knowledge within this cross-disciplinary field in order to understand thecomplex processes during ground work that have detrimental consequences for foundation pile durability. Weconclude that soil types, ground water levels and water flow, as well as the hydrogeological matrix are importantparameters for understanding the stability of a site. These factors will affect the oxygen content available fordevelopment of more intense decay by erosion bacteria or additional attacks by aggressive wood rotting fungi.These interactions and thresholds should be studied further. Knowledge of local environments can determine thevulnerability of burial sites prior to disturbance, and enables taking preventative measures to counter effectsthereof.

1. Introduction

1.1. Use of piling in history

Piling has been used as a building technique for over seven mil-lennia, as known from the UNESCO world heritage site of the pre-historic pile dwellings in the Alpine region dating from as early as 5000BC (Heumller, 2012). Piling was also used extensively by the Romansand was written about by Vitruvius in his famous treatise De architecturalibri decem (“The ten books on architecture”, ca 30 BC), where he citesthe importance of solid ground for laying foundations and describeshow “If solid ground cannot be found […] it must be set with piles […]and the foundations laid on them in the most solid form of construc-tion” (Vitruvius et al., 1914). The building technique is performed byhammering down several piles vertically into the soft soil at evenlyspaced intervals, usually around 10–30 cm in diameter and often sev-eral meters long (Klaassen, 2008), to support the overlying foundationsand transfer the load deeper into the soil. This can be done either usingshort piles (1–3 m long) set close together to compact the soil, or usinglonger, thicker piles (10–15 m long) that reach solid bedrock beneathoverlying soft soil deposits (Przewłócki et al., 2005). There are alsostructures built where the bedrock is inclined, for which part of the

foundation rests on piles that reach a lower firm substrate and otherparts are supported by closely set short piles. Timber logs of softwoodspecies such as pine, spruce and larch were the standard material forfoundation piles up until the middle of the twentieth century (Klaassenand Creemers, 2012), sometimes sharpened in one end to facilitatehammering into the soil.

Solidifying the ground beneath foundations in this manner meantthat even areas with poor soil quality could be developed for housing.Land reclamation of marshland and wasteland colonization has beenused as a method to expand territory since medieval times (Curtis andCampopiano, 2014), so piling came to be a common stabilizing tech-nique throughout Europe and the world. Today, the materials used aredifferent but the method is still relevant as expanding cities need todevelop land reclaimed from the sea or utilize deep foundations tosupport high-rise buildings to save surface space (Jiao et al., 2006).

In waterlogged soils the wooden piles have a long service life andtherefore many historical cities close to rivers, waterfronts and the seaare still standing on original wooden pilings. Several of these old townswith well-preserved historical buildings are today popular tourist at-tractions, such as Stockholm, Trondheim, and Amsterdam, and are animportant part of our shared cultural heritage. This makes the study ofwood decay in urban environments a cultural heritage issue as well as

https://doi.org/10.1016/j.ibiod.2020.104913Received 13 September 2019; Received in revised form 16 January 2020; Accepted 22 January 2020

∗ Corresponding author.E-mail addresses: johanna.elam@marine.gu.se (J. Elam), charlotte.bjordal@marine.gu.se (C. Björdal).

International Biodeterioration & Biodegradation 148 (2020) 104913

0964-8305/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).

T

an engineering issue (Klaassen and Creemers, 2012). In Fig. 1, a 100-year-old pile is taken for analysis outside the parliament building inStockholm. Venice, also an historic cultural heritage city built withpiling, does not face the same challenges of groundwater level changesdue to infrastructural groundwork. However, we have included casestudies from this area as these studies give important insights intobacterial degradation of wooden piles.

1.2. Wood as a sustainable material

Wood is a strong natural composite consisting of small elongatedhollow fibres mainly oriented along the stem axis. These are seasonallyformed in the cambium, just beneath the bark, and are closely con-nected to each other by anatomical openings (pits) in the fibre cell wallwhich secure an even distribution of water, nutrients and extractives inthe living tree (see Supplementary Material Fig. 1). Chemically, the cellwall consists of the polymers cellulose, hemicelluloses, and lignin,which are arranged in micro- and macro-fibrils (Bowyer et al., 2007).The outer part of the stem, named sapwood, transports water in theliving tree whereas the inner part, named heartwood, is the mature non-conductive part of the stem which is rich in extractives. Extractivesoften contain natural pesticides and fungicides which generally makesheartwood more resistant to biological attack. However, like all organicmaterial, wood may undergo a full decomposition by macro and mi-croorganisms in nature (Eaton and Hale, 1993). Above ground and inaccess to oxygen and moisture, wood is rapidly degraded by aggressivefungi, whereas wood in waterlogged and anaerobic soils is highlyprotected (Blanchette et al., 1990).

From studies on wooden foundation piles, archaeological objectsand historic constructions buried in deep anaerobic waterlogged soil,slow bacterial attack on the woody cell wall is the only type of de-gradation that occurs here and this process may take up to hundreds ofyears in a stable environment (Kim et al., 1996; Björdal et al., 2000;Blanchette, 2000; Gregory et al., 2002; Macchioni et al., 2013). A large

threat to stability of foundation piles will arise if the protective wa-terlogged environment changes and become more oxygenated.

1.3. Problems for urban infrastructural groundwork

Modern society places an increasing demand on urban infra-structure to house and transport growing populations, leading to ex-panding urbanization as well as development in and around existingconstructions and buildings. It is estimated that more than half of theworld's population will be living in urban areas by the year 2030(Palmer et al., 2004). Since surface space is limited, this puts a stress onthe urban underground environments. This is especially true during theconstruction period, affecting the local subsurface soil environment atthe construction site as the soil structure is disrupted and great amountsof soil are removed. A common effect of an active construction site is alocal lowering of the groundwater table, e.g. when excavating a tunnelor laying deep foundations, by pumping away the groundwater thatseeps out through the exposed subsoil at the sides of the dig site or intothe tunnel. Although temporary dig sites mostly have a temporary effecton local groundwater levels, tunnels often have persisting groundwaterleakage and therefore contribute to a long-term lowering of ground-water levels. It is common to screen off the area of an excavation sitewith vertical sheet metal walls inserted into the soil as a curtain, toavoid water flooding the dug hole and with the aim to keep the areaoutside the dig site untouched and unaffected, but the technique has itslimits (Wang et al., 2016). Some studies have described situationswhere buildings in such areas may suffer from settling although it mighttake several years before it becomes evident (López Gayarre et al.,2010).

Keeping the environment stable is key for the function of a piledfoundation, since the stability of piled foundations depend not only onthe wooden piles but rather on the interactive system of woodenfoundation piles, surrounding soil matrix and groundwater. To functionas intended, all three components must be present. Infrastructure workscan alter the environment locally and thus affect the function of thefoundation. If the groundwater level is lowered it will lead to desicca-tion of the soil and give rise to a range of problems for the integrity andstrength of the wooden foundation piles. When air and oxygen are in-troduced into the soil and reach the wooden foundation piles, a moreaggressive biodegradation by fungi and bacteria will develop (Eatonand Hale, 1993). This will mainly cause damage to the pile heads, i.e.the part closest to the surface and in contact with the foundation above,leaving the majority of the pile length intact as the deeper buried ma-terial remains waterlogged and outside the range of the decay organ-isms. As the degradation weakens the material, the foundation piles willbecome unable to support their designed loads and the foundation pileswill splinter and fail. Damages are most common where there is anuneven settling rate across the foundation and differential settling oc-curs. This can be due to different settling rates between soil and thepiled foundation, or due to failure of single piles in a foundation. Thepart of the foundation that overlies failed piles will become un-supported, consequently leading to local settling damages and cracks inthe building (see Fig. 2). In severe cases the entire building can collapseas settling damages can be devastating for the structure and the pre-servation of cultural heritage buildings.

Since the foundations are hidden from view, decay can proceed for along time before visible damages are evident in the overlying structure,such as cracks in the façade or a tilt caused by uneven settling rates(Klaassen, 2008; Klaassen and Creemers, 2012). The long timescalesinvolved make it difficult to determine what initiated the decay, andwho is financially responsible for the damages: the owner of thebuilding, or contractors that have performed recent groundwork in thearea (González-Nicieza et al., 2008).

Fig. 1. A 100-year-old pile is excavated from outside the parliament building inStockholm for analysis to determine decay.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

2

1.4. Aim

With this cross-disciplinary paper we aim to provide a compre-hensive overview of main factors that affect the long-term stability ofsubmerged wooden foundation piles in urban environments exposed tolarge-scale construction work. First, we describe urban soils and definethe main factors that could affect the decay rate of wood immersed inthese environments. Secondly, we provide an introduction to the mi-crobial organisms responsible for wood decay in wet soils, and outlineand summarize the biogeochemical factors known to influence the rateof bacterial degradation. Some experiences from previous case studiesare given. Finally, we synthetize current knowledge from the differentfields, and provide new perspectives as well as highlight areas wherefurther studies are necessary in order to understand and, in the future,better control the environment at sites affected by construction work.Since the focus lies on protection of existing wooden piles, preservativetreated piles as well as different piling techniques or coating methodsfor reducing soil friction stress are not part of the study.

This work aims to be a bridge between disciplines and invite newperspectives. As such, we have chosen to treat all subject areas with thesame level of detail and will not present e.g. stoichiometric relation-ships or specific biogeochemical processes in the soil. Our focus is noton the processes themselves but rather how these affect the degradationof wood. Presently, very little is known on the relationship betweenburial environment, site-specific soil processes and the rate of wooddecay. With this work, we hope to inspire novel research ideas and tofill out this knowledge gap.

2. Urban site soil characteristics

2.1. Physical and chemical nature of urban soils

“Urban soils” is an umbrella term for all soils that have been “pro-foundly affected by urbanization”, or soils that have been altered fromtheir natural state by human activity (Lehmann and Stahr, 2007). Herewe use the term in the meaning anthropogenic inner-urban soils, dis-turbed soils that exist within the boundaries of a city. Urban soils arevariable in their composition depending on the city age and locationbut share common characteristics of having been changed from theirnatural state by human activity, such as mixing soils from different

depths and areas, filling natural troughs with foreign soil to level out orraise the ground, and/or contaminating the soil with foreign com-pounds. Not uncommonly, urban soils have lithological discontinuitiesfrom these activities, with older soils overlying younger soils andabrupt changes between one layer and the next, creating interfacesbetween soils of different origin (Craul, 1985). Older cities can alsohave deposits of historic organic material underlying layers of soil andnew buildings, known as cultural layers.

The uppermost layers of urban soils are often compacted, or com-pressed, due to amongst other things destroyed soil structure and heavysurface traffic. Compacted soils contain fewer pore spaces, which re-stricts both water infiltration and gaseous diffusion (Craul, 1985). Thesoil is often less hospitable to soil organism activity such as seedlingemergence and plant root growth when compacted (Awadhwal andThierstein, 1985), which further hinders both the aeration and waterdrainage of the subsoil underlying the compacted surface crusting.

It is common for urban soils to be slightly alkaline and the increasein pH is considered to be due to the weathering of construction mate-rials and the application of de-icing products on roads and pavements(Lehmann and Stahr, 2007). Urban soils in general have elevated levelsof organic carbon, nutrients and contaminants from combustion re-sidues and atmospheric deposition of aerosols (Craul, 1985). Thoughthere are plenty of nutrients accessible, the soil nutrient cycling is in-terrupted by the lack of soil organisms and restricted water movement.The soil has variations in both moisture content and nutrient con-centration, with areas of both excess and limited supply.

Soil temperatures are comparatively higher in urban environmentsthan in surrounding rural environments, as the compacted surface crustacts as an insulator and limits the heat exchange with the atmosphere(Fokaides et al., 2016). Buildings have been seen to have a warmingeffect on the soil, as soils underlying buildings were found to be warmerthan the annual mean air temperature. However, the annual variabilityin soil temperature, oxygen content and water content all decrease withdepth, with the top two to three metres being the most variable(Hollesen and Matthiesen, 2015).

2.2. The urban water cycle

Many urban surfaces such as roads and roofs are covered with im-pervious materials such as concrete, asphalt and sheet metal. Due tocompaction, uncovered urban soil surfaces are also generally lesspermeable than natural or agricultural surfaces. When the precipitationrate exceeds the infiltration rate of the ground in a given area the rainwill exit the system as surface runoff, i.e. water that flows aboveground(Fetter, 2001). Urban environments commonly use drain systems totransport the excess surface water to natural waterways, further redu-cing local infiltration and groundwater recharge. Precipitation on roofshas largely been seen to contribute to surface water drainage systems,but there are systems which enables the roof runoff to feed into theground for local soil moisture and groundwater recharge (Redfern et al.,2016). Runoff water from urban areas contain a higher concentration ofpollutants, suspended particles and organic nutrients than forested oragricultural watersheds, attributable to the inability of urban soils toinfiltrate water to filter out and absorb solutes (Yang and Zhang, 2011).

2.3. Ground water levels

Water that is stored in the saturated layers of soil is known asgroundwater, as opposed to surface water such as lakes or streams. Thelevel in the soil where the water pressure equals the atmosphericpressure is called the water table, comparable to the surface level of alake. Due to capillary forces the soil can be fully saturated even abovethis point, and there can still be soil moisture in the unsaturated zoneabove this level which can drain down through the soil due to gravityand contribute to groundwater recharge (Fetter, 2001).

Groundwater level depends on input and withdrawal from the

Fig. 2. Local settling damages followed up by local adjustments and repair inthe façade of a building due to foundation pile failure that has taken placeunderground.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

3

system and naturally varies over time, due to e.g. seasonal variations ininfiltration rates. In a steady state system where the runoff equals theinput, the water table will stay constant. If the input is reduced then therunoff will slowly drain the system, even if no added withdrawal ismade. These natural variations occur within a normal range determinedby input and outflow, which is different in different environments.Altitude, local precipitation rates and soil type influence the naturalgroundwater level: an organic-rich valley in an area with high pre-cipitation rates will have a higher water table than a sandy plateau withless precipitation (Fetter, 2001). However, human influence can incur adecrease in groundwater level, either directly by e.g. draining wetlandsfor peat production (Van Seters and Price, 2001) or withdrawinggroundwater for use as a fresh water resource (Palmer et al., 2004), orindirectly by reducing the refill rate from natural infiltration of rain-water and altering subsurface water flow paths (Fetter, 2001).

Due to the layered structure of urban soils, it is possible for severalgroundwater reserves (aquifers) to exist in the different layers.Therefore, urban subsurface groundwork can affect the groundwaterlevels in several layers, depending on the location (Fig. 3).

2.4. Water transport in soils

Water always flows from high pressure to low, e.g. with gravity orlaterally from input source to outflow. The hydraulic conductivity of asoil, or the soil's ability to transport water, depends on the particle sizeof the soil, the mean pore size distribution and the homogeneity of thesoils. Generally speaking, a soil consisting of larger particles will con-tain more pore spaces for water to pass through and will thereforetransport water faster than a soil with smaller soil particles. Likewise,the speed at which a soil dewaters depends on the hydraulic con-ductivity, where a soil with high conductivity will drain faster andbecome aerated more quickly than a less conductive soil (Fetter, 2001).

In urban soils, the transport of water including solutes and nutrientsfrom one area to another may be hindered by the altered soil structure.Previous ground work where filling materials of different characterhave been used creates heterogeneity in the ground, which makes itdifficult to predict water flow. Furthermore, underground constructionssuch as building foundations, parking garages and sewers can obstructflow paths, connect groundwater reserves or leak water into the system.Leakage from sewers have been shown to be an important source ofgroundwater, albeit contaminated (Schirmer et al., 2013), and leakingwater mains have been seen to contribute to groundwater recharge in aslarge extent as irrigation (Lerner, 1990). This water is often consideredto contain more oxygen than naturally occurring stagnant groundwater,but the effect of dissolved oxygen content in infiltrating water is de-bated (Matthiesen and Hollesen, 2012).

3. Wood decay in soils

3.1. Aerated topsoil

Wood above ground and in ground contact is decomposed relativelyquickly by the action of wood rotting fungi (white rot, brown rot, softrot) and tunnelling bacteria (Blanchette et al., 1990). These decay formsare well known in the topsoil, and their activity and decay rates areoften described and assessed in studies where preservative impregnatedwood stakes and non-impregnated controls are tested in field experi-ments (Eaton and Hale, 1993). One field experiment showed thatcontrol samples in ground contact were infected to a varying degreeduring the first year, and after four years all stakes were classified as“failures” according to standardized evaluation (Brischke et al., 2014).Similar results were found in other field tests where stakes were halfinserted in soils (Mohebby and Militz, 2010), indicating that wooddegradation in aerated topsoil develops very fast. When durability ofsound and non-preservative treated pine stakes was tested in differentnatural soils, it was concluded that sites often have different site-spe-cific dominating decay profiles where some are more aggressive thanothers (Edlund, 1998; Rapp et al., 2007).

3.2. Waterlogged anaerobic soils

As foundation piles are inserted deep into the ground where theenvironment is anaerobic and waterlogged, the degradation processdiffers from the topsoil environment. In waterlogged environments onlyso-called erosion bacteria (EB) are found able to degrade wood (Kimand Singh, 2000; Björdal, 2012). Analyses of foundation piles fromdifferent geographical locations have all reported on varying degree ofEB deterioration in the wood surface layers (Boutelje and Bravery,1968; Boutelje and Göransson, 1975; Grinda, 1997; Klaassen, 2008;Rehbein et al., 2013). The decay rate is generally extremely slow andtherefore foundation piles can be in service for centuries. Archae-ological wood material witness to still ongoing EB decay, e.g. in Vikingpiles submerged for 1000 years in marine sediments (Björdal, 2000). Ina similar way, foundation piles are most likely continuously degradedby EB during their exposure in ground.

EB are 1–8 μm long rod-shaped bacteria with a diameter of about1 μm (Holt and Jones, 1983; Singh et al., 1990; Nilsson and Daniel,1997). Their way of degrading the wood cell walls and the decay pat-tern left behind are well studied, but their genetic identities are stillunknown (Schmidt and Liese, 1994; Landy et al., 2008). Decay startsfrom the wood surface and proceeds slowly inwards via anatomicalopenings like rays, pits and/or vessels. Here they grow, divide andspread into neighbouring fibres. A typical decay feature of EB is thepreferential decay of the cellulose-rich cell wall, whereas the lignin-rich

Fig. 3. A schematic example of layered soil structurein Gothenburg, Sweden, with modern filling (A), clay(B), coarse-grained mineral soil (C) and bedrock (D).Groundwater stores can exist in a lower aquifer infracture zones in bedrock and the coarse soil, and inan upper aquifer in the uppermost layer of modernfilling. Subsurface groundwork may affect ground-water levels in different layers of the soil, and can bemonitored in wells installed in the lower (H) or upper(I) aquifer. Lowering the groundwater level in eitherof these aquifers affects wooden foundations, by de-siccation of the soil, ground settling or both.Foundation piles that reach bedrock (G) are lesssensitive to ground settling than foundations withoutpiles (E) or with piles inserted in clay (F), but areequally vulnerable to decay in oxygenated environ-ments.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

4

middle lamella surrounding the fibres remain intact (Fig. 4). Duringdecay, residual material is left behind in the hollow centre of the cell(the lumen), and a honeycomb-like skeleton of the middle lamellaepreserves the physical integrity of the wood as long it remains water-logged. In this state, the wood is very weak, soft and spongy, but gives afalse impression of being a well-preserved wood material as the out-ward appearance and original colour are intact (Björdal, 2000).

The bacterial decay alters the chemical composition of the wood,changing the mechanical properties of the wood in proportion. Whenthe strength-providing cellulose is degraded, the density of the wooddecreases and the water content increases, affecting both hardness andstrength of the wood (Capretti et al., 2008).

4. Factors that affect the durability and lifespan of foundationpiles

4.1. Drainage and aeration of soil

Draining soils and lowering the groundwater table can induce set-tling, a lowering of the ground surface due to compaction of the soil, assoil pore spaces that were previously filled with water are compressed.The air that fills the pore spaces when the water drains away is morecompressible than water and is therefore less able to sustain the pres-sure of overlying construction (Budhu and Adiyaman, 2010). Thiscompaction is often an irreversible process since the soil rearrangesduring drying and decreases in porosity, leaving fewer pore spaces forthe water to fill. Organic-rich soils and soils with larger particle size aremore prone to volume loss due to drying (Peng and Horn, 2007).

As the water table recedes, the soil becomes aerated and oxygenpenetrates deeper into the ground. This can cause organic material inculture layers of the soil to decompose, further aggravating the landsettling (Matthiesen, 2007).

Oxygen is regarded as one of the key factors for controlling wooddecay rate in ground. In a fully water saturated environment, the dis-solved oxygen concentrations in the pore water is near anaerobic, andsince fungi demand more oxygen to maintain activity only EB are ableto degrade wood. The decreased activity by EB observed along thedepth gradient are suggested to be attributed the decrease of oxygenconcentrations in the pore water system (Björdal et al., 2000; Hollesenand Matthiesen, 2015), but when oxygen is totally excluded and theenvironment is regarded as anaerobic, decay still takes place

(Kretschmar et al., 2008a).

4.2. Soil pore water and nutrients

Soil pore water can contain not only dissolved oxygen but also,depending on the ion composition, oxygenated ion species which can beused as alternative electron acceptors during organic matter degrada-tion under anaerobic conditions. There are microorganisms that canmake use of the oxygen bound in ions and changes in pH can lead to thedissolution of soluble minerals in the soil. Soils have a natural bufferingeffect which can help protect against pH fluctuations, and while bothorganic and clay components are known to increase the buffering ca-pacity of a soil (Weaver et al., 2004), care should be taken to make sureartificially infiltrated water and filling material match the chemical andphysical properties at the dig sites before disturbance to minimize da-mages caused by a variable environment (Caple, 2004).

It has been suggested that increased levels of nutrients associated toa waterlogged site in Stockholm could explain the severe bacterial de-gradation observed in the piles (Boutelje and Göransson, 1975). But in alarger European study, no positive correlation was found between nu-trient loading and bacterial growth (Huisman et al., 2008a; Kretschmaret al., 2008b). Contrarily, laboratory microcosms indicated that lowernitrogen concentrations promoted decay (Kretschmar et al., 2008a).

4.3. Soil type and physical environment

Foundation piles are mainly inserted in soft soils that are prone tosettling; waterfront clays or muds, marshland peat or historic culturallayers of older cities. Clays and muds are fine-grained soils, such asalluvial deposits from the previous ice age or soils comprised of pre-vious sediment, consisting of fine particles of unconsolidated materialwith small pore spaces. These types of soil have a low hydraulic con-ductivity and are slow to transport water. Some clays can containchemically bound water and can physically swell in contact with waterdepending on the ionic composition of the clay mineral (Hensen andSmit, 2002). Organic-rich soils on the other hand drain faster, as the soilis heterogeneous in terms of grain size, pore size, and permeabilitysince the soils often contain different types of mineral and organicmaterials, such as roots, fungal hyphae and particular organic matter(Jastrow, 1996).

Different soil types have different water retention capacity, orability to store water. There are several forces acting on soil water thatcan act to draw water upwards and maintain moisture in the soil abovethe groundwater table, such as capillary rise due to surface tension,cohesion of water to water and adhesion of water to soil (Fetter, 2001).These forces can help to counter the effects of sinking groundwaterlevels and keep objects in the soil waterlogged. Studying the micro-morphology of a burial medium can give information on water contentof soils overlying the water table, illuminate active soil processes andindicate their spatial scale (Huisman, 2007).

Desiccation not only promotes increased microbial activity, but alsoincreases the friction stress that acts on the piles from the unsaturatedburial medium (López Gayarre et al., 2010). This stress, also known as“skin friction” or “shaft resistance”, reduces the load bearing capacity ofthe pile as the movements of the shrinking soil exerts force on the pile(Fellenius, 1972). Studies on the dynamic load-bearing characteristicsof concrete foundation piles show that the moisture content of the soilmatrix surrounding the foundation piles affect the piles’ capacity towithstand stresses due to soil movement. In a saturated state, the soilbecomes more elastic and can move with the foundation pile, whereas aless saturated soil is stiffer. When exposed to stress, such as vibrationsfrom heavy traffic, a saturated soil will shift and reshape around thepile and support the foundation pile evenly (Aubakirov, 1975).

Increased temperatures are known to promote bacterial activity insoils (Díaz-Raviña et al., 1994), and seasonal variations have been ob-served in the upper layers of wetland soil, with greater abundance and

Fig. 4. Cross section of wood tissue from degraded foundation pile (softwood)showing typical erosion bacteria decay of the wood fibres. Most cell walls arefully degraded (blue) and only a few are still sound and non-degraded (white,arrows). The intact lignin-rich middle encloses the fibres and preserves thephysical integrity in waterlogged state.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

5

respiration during the warmer spring and summer months. As soiltemperature decreases with depth and becomes more stable and con-stant, the amount of bacteria and their activities also decreases re-gardless of the nutrient availability (Douterelo et al., 2009).

5. Selected case studies on piles and archaeological wood

5.1. Case studies on foundation piles

During the 20th century a large number of foundation piles havebeen examined in connection with observed settlings and instability ofbuildings around the world, but only a few reports reached the scien-tific community. In 1968, Boutelje and Bravery described that testing offoundation piles in Stockholm had been ongoing for 20 years as infra-structural urban construction work had changed the ground water le-vels and raised a concern for the condition of the piles (Boutelje andBravery, 1968). In their paper, results on morphology and strength oftwo pine and two spruce piles were given and decay of bacteria-likeorigin, not fungal, was observed in the surface layers. Since then, de-scriptions of EB decay patterns in piles have been reported. Sometimesthese reports include information on correlated strength losses(Lundström, 1981; Paajanen and Viitanen, 1988; Grinda, 1997), che-mical analyses (Gelbrich et al., 2008; Macchioni et al., 2016), or both(Boutelje and Göransson, 1975).

The most complete case study on foundation piles to date is prob-ably the cross-disciplinary research project BACPOLES funded by theEuropean Commission. The full title of the project is “Preserving cul-tural heritage by preventing bacterial decay of wood in foundation pilesand archaeological sites”, and it took place in 2002–2005. In total 27field sites in 6 countries were examined, of which 13 were sites withfoundation piles of pine, spruce, fir, poplar, and oak. These were stu-died in their burial context and the degradation was assessed afterburial for 100–2500 years. All piles showed degradation to a varyingdegree, almost exclusively by erosion bacteria (EB), but the heartwoodoften remained sound and un-degraded. The relative lignin contentincreased as the bacteria consumed the polysaccharides (Gelbrich et al.,2008). Despite in-depth statistical analyses, it was not possible to cor-relate decay rates to specific chemical or physical factors in the en-vironment, although certain sites seemed more aggressive than others.A low redox potential of the soil did however indicate a more protectiveenvironment (Huisman et al., 2008b). Complementary laboratory ex-periments showed that increased nitrogen concentration in the sur-rounding soil matrix seemed to inhibit EB decay (Gelbrich et al., 2008;Kretschmar et al., 2008b). As the identification of EB was a major focusof the project, DNA analyses were given much effort. Results revealed ahuge amount of bacterial diversity inside the wood, but it was notpossible to identify EB within this mixed population. However, withhelp of specially developed isolation and cultivation methods and op-tical tweezer techniques, almost pure cultures indicated that EB couldbelong to the groups Cytophaga or Flavobacteria (Landy et al., 2008;Nilsson et al., 2008).

When full-length piles were examined, variations in degree of decayalong the stem could be evaluated. From a large study in theNetherlands, a majority of 57 full length piles showed a clear decreasein decay with increasing depth in soil (Klaassen and Creemers, 2012).Similar results were found in 3 poles from Berlin (Grinda, 1997) andwhen piles from Venice were examined it was evident that longer pilesshowed a clear depth-gradient in decay whereas shorter piles of about1 m were evenly degraded (Macchioni et al., 2016). An evenly dis-tributed degradation was also found in two 7 m long piles fromStockholm. Here piles were inserted in soil on a small island, whereoxygenated streaming water constantly passed by, which could explainthe homogeneous decay along the full pile length (Berglund et al.,2006).

A more recent study conducted in Riga, Latvia, explored the de-gradation of foundation piles supporting the 13th century Riga

Cathedral. It was concluded that erosion bacteria were the main de-graders of the wooden foundation piles and future threats to thefoundation stability were identified as where the wood was subject to afluctuating groundwater level (Irbe et al., 2019).

Forensic analysis of collapsed foundation piles and the surroundingenvironments are rare. In two studies in Spain, pile failures with de-vastating consequences for the building above revealed a strong cor-relation to previous drainage and construction work in the neighbour-hood. These changed environmental conditions increased the decayprocess until the piles could no longer bear their designed loads andcollapsed. Experience from these events have made the authors suggestand indicate a number of conditions in which the strength of a foun-dation is at risk, such as lowered groundwater table, land subsidenceand friction stress (González-Nicieza et al., 2008; López Gayarre et al.,2010).

The type of foundation is also important to consider, as a numericalstudy from Venice shows that the magnitude of settlement is higher infoundations with short piles than in foundations that reach a lowerstable substrate. However, the short-piled foundations were less likelyto suffer from differential settling. Both of these results could be due tothe fact that completely degraded piles can still provide support as thefriction between piles is still present (Bettiol et al., 2014). As the wooddegrades, the stresses are transferred from the wooden piles to thesurrounding soil matrix instead. The overlying structure is then de-pendent on the soil settling rate (Ceccato et al., 2014).

5.2. Case studies on archaeological wood

There is a wealth of knowledge to be found in archaeological re-search highly relevant to degradation of wooden foundation piles inurban environments. Sometimes large volumes of wood are foundpreserved in waterlogged environments as remaining historic evidenceof houses, bridges, roads, and trackways, but delicate objects of muchsmaller dimensions are also found at sacrificial sites. A few of theseimportant rural or urban archaeological sites have been carefully ex-amined in order to understand the ongoing degradation processes, andto evaluate if the site has a future potential for long term storage of thefind – an option called in-situ or reburial in site preservation.

The Rose Theatre in London, UK (remains of the old ShakespeareTheatre), was one of the first urban archaeological sites being scienti-fically investigated and reburied in situ. In the paper The Rose Theatre:Twenty Years of Continuous Monitoring, Lessons, and Legacy (Corfield,2012) the background and decisions for site management and publicdisplay of The Rose Theatre are explained. Here, a review on in situpreservation is also given in a historic perspective with focus on north-western Europe. It is emphasized that knowledge on the geophysicaland chemical characteristics of the site itself is crucial in order to tailor-make a long-term protective environment with high water levels andanoxic water.

The urban archaeological deposits beneath the old town Bryggen,Bergen, Norway, is another example where site parameters were mea-sured over a long period and analysed in order to establish a moreprotective environment around the organic remains. It was decided topreserve and secure the site by raising the groundwater level and re-ducing the oxygen concentration of the ambient water (Rytter andSchonhowd, 2015). A new environmentally sustainable approach wasdeveloped where rainwater was used as infiltration water and de-oxy-genated in soil beds before spreading into the cultural layers (de Beeret al., 2016).

In rural environments, drainage of wetland and peat bogs oftenthreats the stability and survival of organic archaeological remains.When the water level decreases, desiccation starts and allows air withoxygen to penetrate into previously anaerobic and protective soil layerswhere the archaeological remains are situated. Our understanding ofthese environments and degradation processes started with the in-vestigation of individual sites.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

6

One early example is the sacrificial peat bog Nydam in Denmark,containing a large number of delicate artefacts in wood and metal fromthe Iron Age. The wooden finds were situated relatively close to the soilsurface and heavily degraded by mainly EB. A monitoring program wascarried out to understand the geophysical and chemical parameters aswell as variations and fluctuations in the area. Results from monitoringreveal a still anoxic waterlogged site with high water level and oxygenonly in the upper centimetres of the bog (Gregory and Matthiesen,2012). Today monitoring is still ongoing and modern sound woodsamples are inserted as proxy indicators for periodic analyses (Gregoryet al., 2008).

At two archaeological sites in the UK, large interventions have beencarried out to preserve historic constructions in wetlands from de-siccation. Observation of drainage in the area of the 2 km long Neolithicwooden trackway Sweet Track led to decisions on artificial water levelraise to limit oxygen supply (Brunning et al., 2000).

At the Bronze Age site Flag Fen, results from environmental mon-itoring and examination of archaeological wood and inserted soundwood stakes indicated a microbial aggressive site with sinking waterlevels (Powell et al., 2001). Thus, an artificial lake with increased waterlevels was constructed in 1987 to flood the area. The condition of thesite today is not known (Malim et al., 2015) although hydrogeologicalmodelling indicates that many timbers are still situated in non-pro-tected soil layers (Wagstaff et al., 2016).

5.3. Summary case studies

The lesson learned from case studies in terrestrial waterlogged en-vironments, both on foundation piles and on archaeological material, isthat the rate of degradation in wood depends on many factors, whichinteract in a complex manner that is not yet fully understood.

Many case studies point out single physical and chemical para-meters that affect decay rates of organic material in general, such as soiltype, depth of burial, nutrients, oxygen availability, ground water levelsand water flow. Redox potential and pH give important information andvalues on the aggressiveness of complex interactions (Caple, 1994;Caple et al., 1996; Smit, 2004; Huisman et al., 2008a). In addition,factors related to the wood material itself such as wood species, dur-ability, amount sapwood and heartwood, age of the wood, number andwidth of annual rings and late wood/early wood contribute to thecomplexity (Klaassen and Creemers, 2012), just as the age of the con-struction and the size/thickness are important variables. For full un-derstanding and managing of a site containing organic material, addi-tional information of the natural fluctuations due to weather should beincluded too (Matthiesen and Hollesen, 2012). As urban environmentsare often not uniform, site-specific groundwater conditions also need tobe taken into consideration, such as at the archaeological site Bryggen,Bergen (Matthiesen, 2008).

6. Conclusions and future perspectives

When planning urban infrastructural projects, it is important toconsider how coming intervention and ground work could affect thesoil environment and potentially shorten the service life of pre-existingwooden foundation piles in the surroundings.

It is evident that a stable and high groundwater level is one of themost important environmental factors for long term durability offoundation piles. Uncontrolled local lowering of groundwater levels aturban construction sites starts a desiccation process in the soil sur-rounding the pile heads. Here, increased oxygen concentrations sti-mulate microbial wood degradation by fungi and bacteria, which willreduce the strength and consequently the service life of the piles. Littleis known on the time scale of these processes; is it in terms of weeks,months, years or decades? When historic pile heads are analysed, ob-served decay status often reveals activity by EB in the surface layers ofthe pile, progressing inwards. Whether these are related to unavoidable

slow decay in nature or accelerated decay caused by anthropogenicimpacts is rather difficult to determine. However, presence of soft rot,white rot or brown rot decay in pile heads implies much faster de-gradation processes strictly related to a more oxygenated environment,which most likely correlates to more recent or contemporary changes atthe site.

Urban soils are often characterized by elevated concentrations oforganic nutrients. Among those are compounds which are known tostimulate many biological processes in soil and water. However, despitethat some archaeological and rural sites have been characterized inregards to soil type and nutrients available, very little is investigated ontheir correlation to wood decay and especially to the speed of bacterialdegradation. Perhaps surprisingly, preliminary results indicate that ni-trogen does not increase decay by EB (Kretschmar et al., 2008a).

The hydrogeological water flow at a site is crucial for a better un-derstanding of the oxygen and nutrient transport around the piles, bothin the upper soil layers as well as in the deeper parts. Here we learnedthat water transport in urban sites is different from rural sites, due tothe fact that impermeable ground surfaces in urban cities exclude aneven distribution of rainwater. Horizontal water flow in conductive soillayers is therefore most important for maintaining a high groundwaterlevel beneath streets and buildings. One factor that might hinder a freehorizontal water flow is the building itself. As several hundreds of pilesmay support the foundation of a large building, set as close as 0.6 mapart (Pallav and Mets, 2013), deep exposed foundations piles mayalter the hydrological behaviour of soft soils by damming the ground-water flow and reducing the hydraulic conductivity (Ding et al., 2008).The piles are often concentrated around the outer edges of a foundationand load-bearing walls, hindering water flow underneath the structureas the sides are blocked by the piles. The current dewatering methodsfor excavations in soft deposits fail to take this into consideration (Maet al., 2014), and assuming a uniform drawdown around the excavationsite would underestimate the effects near the foundation piles.

Another factor that causes uneven distribution of oxygen is theheterogeneous soil matrix, where complex interactions between the soilenvironment and bacterial decay take place. The significance ofstudying wood decay within its site context is highlighted in case stu-dies of both archaeological and urban origin (Holden et al., 2006). Theconditions of the site and the ability of the soil matrix to rebuff changesto the environment are crucial to the survival of the wood. Empiricalanalysis of well-preserved wood specimens can give us an idea of whatconstitutes a good preservation environment, and comparing these withdata from forensic analyses of pile foundation failures could helpidentify environmental conditions that make buried wood more or lesssusceptible to decay. By identifying the main drivers of decay and de-veloping a method to determine the hostility of different environments,predictive models can be developed that allow for protective measuresto limit disturbances before exposed foundation piles are weakened tothe point of failure.

During the management of archaeological sites, active interventionis sometimes used to maintain waterlogged conditions. At the Neolithicwooden trackway “Sweet Track” in Somerset, England, the effects of alocally lowered groundwater table were offset by continuouslypumping water to maintain moisture (Brunning et al., 2000). Similarly,artificial infiltration of tap water is sometimes used to counter de-siccation at urban sites. However, artificially infiltrated water oftencontains more oxygen than the original groundwater and might there-fore increase bacterial decay even if it counters other adverse effects.Further studies are necessary to determine how serious this threat is.There can also be traces of chemicals in the water, either dissolved inthe water at an earlier point (Bucheli et al., 1998) or accumulated at theinfiltration site (Gschwend et al., 1990). Changes to the chemical natureof the soil pore water induced by artificial infiltration water may alsoalter the redox potential of the soil surrounding the wood and maystimulate degradation (Caple, 2004).

Redox potential is often used as an indicator to determine how

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

7

preserving an environment is. A nearly neutral pH and reducing en-vironment co-occurs with anoxic conditions, thereby limiting the ac-tivity of decay organisms. Investigations and analyses from archae-ological sites suggest that a redox potential below −100 mV or −200mV indicates good preservation conditions for organic material (Capleet al., 1996; Jordan, 2001; Matthiesen et al., 2004), but the criticalthreshold values for wood degradation are so far unknown.

In order to accurately predict decay, the relevant timescales forwood decay must be considered and the so-called tipping points iden-tified, i.e. the environmental limits beyond which an interference willcause damages. Furthermore, there is a need for increasing use of sa-crificial samples and development of proxy indicators for decay, e.g.using typical degradation patterns to approximate the state of decay(Klaassen, 2008), as using entire foundations as test samples is un-sustainable. Both field studies with samples of naturally decayed woodtaken in situ as well as laboratory experiments with sound wood cangive crucial insight into decay processes and environmental factors thateither inhibit or promote bacterial decay (Gregory et al., 2008). Com-bining field observations with monitored lab experiments could give anadded value, as the field studies provide a baseline and microcosmstudies can simulate future conditions, such as altered groundwaterlevels or increased temperatures due to climate change or local dis-turbance from infrastructure projects (Kretschmar et al., 2008a).

It is evident that there are many factors that influence the rate ofdecay in urban environments, and there is still much to learn in eachseparate field of knowledge. However, there exists a complex and dy-namic relationship between different factors that influence each other.Only a few studies have contributed to present knowledge and it is clearthat this broad and interdisciplinary field of research needs more at-tention. Future research should include improving methods for datacollection, as there is an increasing need for both better monitoring andmodelling data, as well as specific research into wooden foundationpiles for development of guidelines. Since urban environments havedifferent characteristics and biogeochemical cycling regimes from theirrural counterparts, it is necessary to conduct specific studies of wooddecay in this setting which could give further insight into how well ahistoric foundation is equipped to withstand attack.

This overview has compiled factors that have an impact on thedegradation of pile foundations in urban environments. Biological,chemical, and hydrogeological factors interact in a complex urbansystem. When studying decay in urban environments, there are there-fore several interlocking aspects that need to be taken into considera-tion. Results from case studies in both field and laboratory settings havehighlighted the importance of studying samples within their site con-text, and for a further understanding of the processes affecting wooddecay we need more detailed data providing us with threshold values,especially those related to oxygen concentrations in diverse water/soilsettings.

Today there is a lack of detailed environmental parameters withthreshold data correlated to wood degradation. Future research andstudies must be designed with the relevant question in mind, so thatguidelines describing protective and passivating environment duringurban construction work can be given.

Declaration of competing interest

None.

Acknowledgements

The authors thank Helle Ploug and Bengt Åhlén for their insightfulcomments that have helped to improve this paper.

This work was funded by the Swedish Transport Administration.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ibiod.2020.104913.

References

Aubakirov, A.T., 1975. Experimental investigations of the dynamic characteristics of pilefoundations. Soil Mech. Found. Eng. 12 (4), 234–237.

Awadhwal, N.K., Thierstein, G.E., 1985. Soil crust and its impact on crop establishment: areview. Soil Tillage Res. 5 (3), 289–302.

Berglund, C., Holm, G., Nilsson, T., Björdal, C.G., 2006. Durability of old timber piles. Acase study. In: Butcher, A.P., Powell, J.J.M., Skinner, H.D. (Eds.), Proceedings ofInternational Conference "Reuse of Foundations for Urban Sites, RuFUS”. IHS BREPress, Berkshire, pp. 79–86.

Bettiol, G., Ceccato, F., Pigouni, A.E., Modena, C., Simonini, P., 2014. Effect on thestructure in elevation of wood deterioration on small-pile foundation: numericalanalyses. Int. J. Architect. Herit. 10 (1), 44–54.

Björdal, C.G., 2000. Waterlogged Archaeological Wood : Biodegradation and itsImplications for Conservation, PhD Thesis. Swedish University of AgriculturalSciences, Uppsala, Sweden.

Björdal, C.G., 2012. Microbial degradation of waterlogged archaeological wood. J. Cult.Herit. 13 (3), S118–S122.

Björdal, C.G., Daniel, G., Nilsson, T., 2000. Depth of burial, an important factor in con-trolling bacterial decay of waterlogged archaeological poles. Int. Biodeterior.Biodegrad. 45 (1–2), 15–26.

Blanchette, R.A., 2000. A review of microbial deterioration found in archaeological woodfrom different environments. Int. Biodeterior. Biodegrad. 46 (3), 189–204.

Blanchette, R.A., Nilsson, T., Daniel, G., Abad, A., 1990. Biological degradation of wood.Adv. Chem. 225, 141–174.

Boutelje, J.B., Bravery, A., 1968. Observations on bacterial attack of piles supporting aStockholm building. J. Inst. Wood Sci. 20, 47–48.

Boutelje, J.B., Göransson, B., 1975. Decay in wood constructions below the ground watertable. Swed. J. Agric. Res. 5, 113–123.

Bowyer, J., Shmulsky, R., Haygreen, J., 2007. Forest Products and Wood Science: anIntroduction, fifth ed. Wiley-Blackwell, Hoboken, NJ.

Brischke, C., Meyer, L., Olberding, S., 2014. Durability of wood exposed inground–Comparative field trials with different soil substrates. Int. Biodeterior.Biodegrad. 86, 108–114.

Brunning, R., Hogan, D., Jones, J., Jones, M., Maltby, E., Robinson, M., Straker, V., 2000.Saving the Sweet Track: the in situ preservation of a neolithic wooden trackway,Somerset, UK. Conserv. Manag. Archaeol. Sites 4 (1), 3–20.

Bucheli, T.D., Müller, S.R., Heberle, S., Schwarzenbach, R.P., 1998. Occurrence and be-havior of pesticides in rainwater, roof runoff, and artificial stormwater infiltration.Environ. Sci. Technol. 32 (22), 3457–3464.

Budhu, M., Adiyaman, I.B., 2010. Mechanics of land subsidence due to groundwaterpumping. Int. J. Numer. Anal. Methods GeoMech. 34 (14), 1459–1478.

Caple, C., 1994. Reburial of waterlogged wood, the problems and potential of this con-servation technique. Int. Biodeterior. Biodegrad. 34 (1), 61–72.

Caple, C., 2004. Towards a benign reburial context: the chemistry of the burial en-vironment. Conserv. Manag. Archaeol. Sites 6 (3–4), 155–165.

Caple, C., Dungworth, D., Clogg, P., 1996. Results of the characterization of the anoxicwaterlogged environments which preserve archaeological organic materials. In:Hoffman, P., Grant, T., Spriggs, J.A. (Eds.), Proceedings of the 6th ICOM Group onWet Organic Archaeological Materials Conference, ICOM Committee forConservation Working Group on Wet Organic Archaeological Materials, pp. 5–71Bremerhaven.

Capretti, C., Macchioni, N., Pizzo, B., Galotta, G., Giachi, G., Giampaola, D., 2008. Thecharacterization of waterlogged archaeological wood: the three Roman ships found inNaples (Italy). Archaeometry 50 (5), 855–876.

Ceccato, F., Simonini, P., Köppl, C., Schweiger, H.F., Tschuchnigg, F., 2014. FE Analysisof degradation effect on the wooden foundations in Venice. Riv. Ital. Geotec. 48,27–37.

Corfield, M., 2012. The Rose Theatre: twenty years of continuous monitoring, lessons, andlegacy. Conserv. Manag. Archaeol. Sites 14 (1–4), 384–396.

Craul, P.J., 1985. A description of urban soils and their desired characteristics. J. Arboric.11 (11), 330–339.

Curtis, D.R., Campopiano, M., 2014. Medieval land reclamation and the creation of newsocieties: comparing Holland and the Po Valley, c.800-c.1500. J. Hist. Geogr. 44,93–108.

de Beer, J., Seither, A., Vorenhout, M., 2016. Effects of a new hydrological barrier on thetemperatures in the organic archaeological remains at Bryggen in Bergen, Norway.Conserv. Manag. Archaeol. Sites 18 (1–3), 99–115.

Díaz-Raviña, M., Frostegård, Å., Bååth, E., 1994. Thymidine, leucine and acetate in-corporation into soil bacterial assemblages at different temperatures. FEMS (Fed. Eur.Microbiol. Soc.) Microbiol. Ecol. 14 (3), 221–231.

Ding, G., Jiao, J.J., Zhang, D., 2008. Modelling study on the impact of deep buildingfoundations on the groundwater system. Hydrol. Process. 22 (12), 1857–1865.

Douterelo, I., Goulder, R., Lillie, M., 2009. Response of the microbial community to watertable variation and nutrient addition and its implications for in situ preservation oforganic archaeological remains in wetland soils. Int. Biodeterior. Biodegrad. 63 (6),795–805.

Eaton, R.A., Hale, M.D.C., 1993. Wood: Decay, Pests and Protection. Chapman and Hall,London.

Edlund, M.L., 1998. Durability of untreated wood exposed in terrestrial test fields andmicrocosms. Mater. Org. 32 (4), 253–275.

Fellenius, B.H., 1972. Down-drag on piles in clay due to negative skin friction. Can.Geotech. J. 9 (4), 323–337.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

8

Fetter, C.W., 2001. Applied Hydrogeology, fourth ed. Prentice Hall, Upper SaddleRiver, NJ.

Fokaides, P.A., Kylili, A., Nicolaou, L., Ioannou, B., 2016. The effect of soil sealing on theurban heat island phenomenon. Indoor Built Environ. 25 (7), 1136–1147.

Gelbrich, J., Mai, C., Militz, H., 2008. Chemical changes in wood degraded by bacteria.Int. Biodeterior. Biodegrad. 61 (1), 24–32.

González-Nicieza, C., Álvarez-Fernández, M.I., Álvarez-Vigil, A.E., Menéndez-Díaz, A.,2008. Forensic analysis of foundation failure in gypsiferous ground. Eng. Fail. Anal.15 (6), 736–754.

Gregory, D., Matthiesen, H., 2012. Nydam Mose: in situ preservation at work. Conserv.Manag. Archaeol. Sites 14 (1–4), 479–486.

Gregory, D., Matthiesen, H., Björdal, C., 2002. In situ preservation of artefacts in Nydammose: studies into environmental monitoring and the deterioration of wooden arte-facts. In: Hoffmann, P., Spriggs, J.A., Grant, T., Cook, C., Recht, A. (Eds.), Proceedingsof the 8th ICOM Group on Wet Organic Archaeological Materials Conference, ICOMCommittee for Conservation Working Group on Wet Organic ArchaeologicalMaterials, pp. 213–223 Bremerhaven.

Gregory, D., Helms, A.C., Matthiesen, H., 2008. The use and deployment of modern woodsamples as a proxy indicator for biogeochemical processes on archaeological sitespreserved in situ in a variety of environments of differing saturation level. Conserv.Manag. Archaeol. Sites 10 (3), 204–222.

Grinda, M., 1997. Some Experiences with Attack of Microorganisms on WoodenConstructions Supporting Foundations of Houses and Bridges. IRG/WP/97 10132.The International Research Group on Wood Preservation, Stockholm.

Gschwend, P.M., Backhus, D.A., MacFarlane, J.K., Page, A.L., 1990. Mobilization of col-loids in groundwater due to infiltration of water at a coal ash disposal site. J. Contam.Hydrol. 6 (4), 307–320.

Hensen, E.J.M., Smit, B., 2002. Why clays swell. J. Phys. Chem. B 106 (49),12664–12667.

Heumller, M., 2012. Erosion and Archaeological Heritage Protection in Lake Constanceand Lake Zurich: the Interreg IV Project Erosion und Denkmalschutz am Bodenseeund Zrichsee. Conserv. Manag. Archaeol. Sites 14 (1/4), 48–59.

Holden, J., West, L.J., Howard, A.J., Maxfield, E., Panter, I., Oxley, J., 2006. Hydrologicalcontrols of in situ preservation of waterlogged archaeological deposits. Earth Sci.Rev. 78 (1), 59–83.

Hollesen, J., Matthiesen, H., 2015. The influence of soil moisture, temperature andoxygen on the oxic decay of organic archaeological deposits. Archaeometry 57 (2),362–377.

Holt, D., Jones, E., 1983. Bacterial degradation of lignified wood cell walls in anaerobicaquatic habitats. Appl. Environ. Microbiol. 46 (3), 722–727.

Huisman, D., 2007. Using soil micromorphology to evaluate the suitability of the burialenvironment for in situ protection. In: Straetkver, K., Huisman, D. (Eds.), Proceedingsof the 10th ICOM Group on Wet Organic Archaeological Materials Conference,Rijksdienst voor Archeologie, Cultuurlandschap en Monumenten, pp. 101–112Amersfoort.

Huisman, D.J., Kretschmar, E.I., Lamersdorf, N., 2008a. Characterising physicochemicalsediment conditions at selected bacterial decayed wooden pile foundation sites in TheNetherlands, Germany, and Italy. Int. Biodeterior. Biodegrad. 61 (1), 117–125.

Huisman, D.J., Manders, M.R., Kretschmar, E.I., Klaassen, R.K.W.M., Lamersdorf, N.,2008b. Burial conditions and wood degradation at archaeological sites in TheNetherlands. Int. Biodeterior. Biodegrad. 61 (1), 33–44.

Irbe, I., Bikovens, O., Fridrihsone, V., Dzenis, M., 2019. Impact of biodeterioration onstructure and composition of waterlogged foundation piles from Riga Cathedral(1211 CE), Latvia. J. Archaeol. Sci.: Reports 23, 196–202.

Jastrow, J.D., 1996. Soil aggregate formation and the accrual of particulate and mineral-associated organic matter. Soil Biol. Biochem. 28 (4), 665–676.

Jiao, J.J., Wang, X.-S., Nandy, S., 2006. Preliminary assessment of the impacts of deepfoundations and land reclamation on groundwater flow in a coastal area in HongKong, China. Hydrogeol. J. 14 (1), 100–114.

Jordan, B.A., 2001. Site characteristics impacting the survival of historic waterloggedwood: a review. Int. Biodeterior. Biodegrad. 47 (1), 47–54.

Kim, Y.S., Singh, A.P., 2000. Micromorphological characteristics of wood biodegradationin wet environments: a review. IAWA J. 21 (2), 135–155.

Kim, Y.S., Singh, A.P., Nilsson, T., 1996. Bacteria as important degraders in waterloggedarchaeological woods. Holzforschung 50 (5), 389–392.

Klaassen, R.K.W.M., 2008. Bacterial decay in wooden foundation piles—patterns andcauses: a study of historical pile foundations in The Netherlands. Int. Biodeterior.Biodegrad. 61 (1), 45–60.

Klaassen, R.K.W.M., Creemers, J.G.M., 2012. Wooden foundation piles and its under-estimated relevance for cultural heritage. J. Cult. Herit. 13 (3), S123–S128.

Kretschmar, E.I., Gelbrich, J., Militz, H., Lamersdorf, N., 2008a. Studying bacterial wooddecay under low oxygen conditions—results of microcosm experiments. Int.Biodeterior. Biodegrad. 61 (1), 69–84.

Kretschmar, E.I., Keijer, H., Nelemans, P., Lamersdorf, N., 2008b. Investigating physi-cochemical sediment conditions at decayed wooden pile foundation sites inAmsterdam. Int. Biodeterior. Biodegrad. 61 (1), 85–95.

Landy, E.T., Mitchell, J.I., Hotchkiss, S., Eaton, R.A., 2008. Bacterial diversity associatedwith archaeological waterlogged wood: ribosomal RNA clone libraries and dena-turing gradient gel electrophoresis (DGGE). Int. Biodeterior. Biodegrad. 61 (1),106–116.

Lehmann, A., Stahr, K., 2007. Nature and significance of anthropogenic urban soils. J.Soils Sediments 7 (4), 247–260.

Lerner, D.N., 1990. Groundwater recharge in urban areas. Atmos. Environ. Part B - UrbanAtmos. 24 (1), 29–33.

López Gayarre, F., González-Nicieza, C., Alvarez-Fernández, M.I., Álvarez-Vigil, A.E.,2010. Forensic analysis of a pile foundation failure. Eng. Fail. Anal. 17 (2), 486–497.

Lundström, R.B., 1981. Quality of the wood in timber piles. In: Publications Committee of

X. ICSMFE (Ed.), Proceedings of the 10th International Conference on Soil Mechanicsand Foundation Engineering. A. A. Balkema, Rptterdam, pp. 131–132.

Ma, L., Xu, Y.-S., Shen, S.-L., Sun, W.-J., 2014. Evaluation of the hydraulic conductivity ofaquifers with piles. Hydrogeol. J. 22 (2), 371–382.

Macchioni, N., Capretti, C., Sozzi, L., Pizzo, B., 2013. Grading the decay of waterloggedarchaeological wood according to anatomical characterisation. The case of the Fiavésite (NE Italy). Int. Biodeterior. Biodegrad. 84, 54–64.

Macchioni, N., Pizzo, B., Capretti, C., 2016. An investigation into preservation of woodfrom Venice foundations. Construct. Build. Mater. 111, 652–661.

Malim, T., Morgan, D., Panter, I., 2015. Suspended preservation: particular preservationconditions within the must farm–flag fen Bronze age landscape. Quat. Int. 368,19–30.

Matthiesen, H., 2007. Preservation Conditions above the Groundwater Level at the Rearof Nordre Bredsgården, Bryggen in Bergen. Results from MB21 and Testpit from2006. National Museum of Denmark, Department of Conservation Report 10832-0011-1.

Matthiesen, H., 2008. Detailed chemical analyses of groundwater as a tool for monitoringurban archaeological deposits: results from Bryggen in Bergen. J. Archaeol. Sci. 35(5), 1378–1388.

Matthiesen, H., Hollesen, J., 2012. Preservation Conditions in Unsaturated UrbanDeposits: Results from Continuous Logging of Oxygen, Water Content andTemperature at the Rear of Nordre Bredsgården, Bergen in the Period October 2010to November 2011. National Museum of Denmark. Department of ConservationReport 11031263.

Matthiesen, H., Gregory, D., Jensen, P., Sørensen, B., 2004. Environmental monitoring atNydam, a waterlogged site with weapon sacrifices from the Danish Iron age. I: acomparison of methods used and results from undisturbed conditions. J. Wetl.Archaeol. 4 (1), 55–74.

Mohebby, B., Militz, H., 2010. Microbial attack of acetylated wood in field soil trials. Int.Biodeterior. Biodegrad. 64 (1), 41–50.

Nilsson, T., Daniel, G., 1997. Developments in the study of soft rot and bacterial decay. In:Bruce, A., Palfreyman, J. (Eds.), Forest Products Biotechnology, first ed. CRC Press,London, pp. 47–72.

Nilsson, T., Björdal, C., Fällman, E., 2008. Culturing erosion bacteria: procedures forobtaining purer cultures and pure strains. Int. Biodeterior. Biodegrad. 61 (1), 17–23.

Paajanen, L., Viitanen, H., 1988. Microbiological Degradation of Wooden Piles in BuildingFoundations. IRG/WP/1370. The International Research Group on WoodPreservation, Stockholm.

Pallav, V., Mets, M., 2013. The history of the foundations in Tartu. Baltic Piling 37.Palmer, M., Bernhardt, E., Chornesky, E., Collins, S., Dobson, A., Duke, C., Gold, B.,

Jacobson, R., Kingsland, S., Kranz, R., Mappin, M., Martinez, M.L., Micheli, F., Morse,J., Pace, M., Pascual, M., Palumbi, S., Reichman, O.J., Simons, A., Townsend, A.,Turner, M., 2004. Ecology for a crowded planet. Science 304 (5675), 1251.

Peng, X., Horn, R., 2007. Anisotropic shrinkage and swelling of some organic and in-organic soils. Eur. J. Soil Sci. 58 (1), 98–107.

Powell, K.L., Pedley, S., Daniel, G., Corfield, M., 2001. Ultrastructural observation ofmicrobial succession and decay of wood buried at a Bronze Age archaeological site.Int. Biodeterior. Biodegrad. 47 (3), 165–173.

Przewłócki, J., Dardzinska, I., Swinianski, J., 2005. Review of historical buildings'foundations. Geotechnique 55 (5), 363–372.

Rapp, A.O., Brischke, C., Welzbacher, C.R., 2007. The influence of different soil substrateson the service life of Scots pine sapwood and oak heartwood in ground contact. WoodMater. Sci. Eng. 2 (1), 15–21.

Redfern, T.W., Macdonald, N., Kjeldsen, T.R., Miller, J.D., Reynard, N., 2016. Currentunderstanding of hydrological processes on common urban surfaces. Prog. Phys.Geogr. 40 (5), 699–713.

Rehbein, M., Koch, G., Schmitt, U., Huckfeldt, T., 2013. Topochemical and transmissionelectron microscopic studies of bacterial decay in pine (Pinus sylvestris L.) harbourfoundation piles. Micron 44, 150–158.

Rytter, J., Schonhowd, I. (Eds.), 2015. Monitoring, Mitigation, Management. TheGroundwater Project-Safeguarding the World Heritage Site of Bryggen in Bergen.Riksantikvaren, Oslo.

Schirmer, M., Leschik, S., Musolff, A., 2013. Current research in urban hydrogeology – areview. Adv. Water Resour. 51, 280–291.

Schmidt, O., Liese, W., 1994. Occurrence and significance of bacteria in wood.Holzforschung 48 (4), 271–277.

Singh, A., Nilsson, T., Daniel, G., 1990. Bacterial attack of Pinus sylvestris wood undernear anaerobic conditions. J. Inst. Wood Sci. 11 (6), 237–249.

Smit, A., 2004. Monitoring the burial environment of terrestrial wet archaeological sitesin The Netherlands. Conserv. Manag. Archaeol. Sites 6 (3–4), 325–332.

Van Seters, T.E., Price, J.S., 2001. The impact of peat harvesting and natural regenerationon the water balance of an abandoned cutover bog, Quebec. Hydrol. Process. 15 (2),233–248.

Vitruvius, P., Morgan, M.H., Warren, H.L., 1914. Vitruvius, the Ten Books onArchitecture.

Wagstaff, S., Cheetham, J., Davis, A., Williams, J., Outram, Z., Priddy, D., Chapman, H.,2016. Hydrogeological modelling of water-level changes in an area of archaeologicalsignificance: a case study from Flag fen, Cambridgeshire, UK. Conserv. Manag.Archaeol. Sites 18 (1–3), 156–169.

Wang, J., Wu, Y., Liu, X., Yang, T., Wang, H., Zhu, Y., 2016. Areal subsidence underpumping well-curtain interaction in subway foundation pit dewatering: conceptualmodel and numerical simulations. Environ. Earth Sci. 75 (3), 1–13.

Weaver, A.R., Kissel, D.E., Chen, F., West, L.T., Adkins, W., Rickman, D., Luvall, J.C.,2004. Mapping soil pH buffering capacity of selected fields in the coastal plain. SoilSci. Soc. Am. J. 68 (2), 662–668.

Yang, J.-L., Zhang, G.-L., 2011. Water infiltration in urban soils and its effects on thequantity and quality of runoff. J. Soils Sediments 11 (5), 751–761.

J. Elam and C. Björdal International Biodeterioration & Biodegradation 148 (2020) 104913

9