&7 Luisa M. S. Borges PhDConsultant, Institute of Marine Sciences, School of Biological Sciences,University of Portsmouth, Portsmouth, UK; L3 Scientific Solutions,Geesthacht, Germany
&8 Graham P. Malyon PhDLead Technician, Institute of Marine Science, School of BiologicalSciences, University of Portsmouth, UK
1 2 3 4 5 6
7 8
Naturally durable species of timber are used as an alternative to preservative-treated timber for marine structures, but
many species have not been evaluated for their potential for use in this environment. BS EN 275 evaluates marine
borer resistance and specifies a 5-year test period: too long a period for screening tests to be economically viable.
Furthermore, the test does not evaluate abrasion resistance. Novel fast-track screening methods were used in this
study to evaluate the marine borer and abrasion resistance of 18 less-used timber species. Comparative resistance was
assessed by comparing the feeding rates of marine borers and abrasion resistance observed in candidate species
against greenheart and ekki, which were used as benchmark species. A number of less-used species, originating from
South America and West Africa, performed better than the benchmark species in laboratory tests and over an
18-month exposure period in the sea. A number of species also performed comparatively well in abrasion trials
although resistance to abrasion does not necessarily correlate with resistance to attack by marine borers.
1. IntroductionThe Environment Agency (the Agency) and other maritimecoastal authorities fund, build and maintain timber structuresin the marine environment throughout the UK, with a particu-lar focus being groynes for control of longshore transport ofbeach material. For such structures, historically, timber hasbeen favoured as a construction material although constructionhas relied on a narrow range of timber species such as
greenheart (Chlorocardium rodiei) and ekki (Lophira alata).The Agency has concerns that this industry sector has becomeover-reliant on the use of these two species and fears that itmay become increasingly difficult to procure these two speciesin sufficient volumes and sizes.
Another challenge is being able to procure these species withsupporting evidence of sustainability and legality that satisfies
1
Construction Materials
Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
its timber-purchasing policy. Illegal logging and unsustainableforest management are now recognised as global problems.Forests are a precious natural resource, and their destructionhas wide-ranging social, economic and environmental impacts.However, there is demand for strong, durable, cost-effectiveand environmentally acceptable construction material (EA,2010a). With this difficulty in mind, the Agency has alsorecognised that alternatives to timber, such as steel, steel-reinforced concrete or rock, for example, may have to be con-sidered for construction schemes (EA, 2010b).
In order to widen their inventory of suitable timber species,the Agency, in collaboration with HR Wallingford and theTimber Research and Development Association (Trada), under-took a research project to assess the potential marine borerand abrasion resistance of candidate less-used species (LUS) oftimber using a range of novel, fast-track laboratory screeningtests, as well as a marine exposure trial. Candidates were iden-tified by desk-based scoping study and previous research(Borges et al., 2003; Crossman and Simm, 2004; Williamset al., 2004a).
Engineers have a key role in making informed decisions on thetype of materials to be used in the coastal schemes they designand construct. The decision must balance technical require-ments with environmental and cost considerations (EA, 2010b;Williams et al., 2004a).
Economics is an important consideration for all coastalschemes. Practically all major capital schemes, such as the cur-rent £17 million groyne-renewal programme at Bournemouth,are dependent on public money. The local authority muststrike a balance between the most suitable material for thejob against funds available. This is even more challenging inthe current climate of austerity where scheme designers mustfind 10% efficiency savings (Harlow, personal communication,2015).
Securing public money for such schemes is a lengthy, com-petitive process. For example, in securing funding for a coastalprotection scheme, the local authority must comply with aShoreline Management Plan and a Strategy Study, which posethe questions ‘what should we do with this management unit?’(e.g. hold the line, retreat the line, advance the line or do-nothing) and ‘how do we do that?’. The project appraisal mustinclude a detailed design of all the options and the economiccase for the works and comply with all formal notices underthe Coast Protection Act 1949 (EA, 2010b). If it is approvedtechnically and economically, the scheme can go ahead(Harlow, personal communication, 2015).
The marine environment is challenging for all constructionmaterials, but timber suffers remarkably little from the effects
of immersion compared with, for example, concrete, which cansuffer from spalling, and steel which can suffer from corrosion(Cragg, 1996).
From a technical point of view, timber retains attractionsover other materials due to its relative durability (Cragg, 1996),resilience, favourable strength-to-weight ratio, and ease of fab-rication and repair. Environmentally, the renewable nature oftimber is attractive, particularly if recycled or obtained fromwell-managed forests (Crossman and Simm, 2004). However,material costs, the proven track record of the timber species,the risk of marine borer attack and the requirement to meetstringent procurement rules may discourage engineers from itsuse (Williams et al., 2004b).
Consequently, this sector of the construction industry is conser-vative, and there is a reluctance to specify timber specieswithout a proven track record. Brazier (1995) summarised thedesirable material properties for timbers used in marine con-struction as being: resistance against marine borers and fungalattack in the marine environment; resistance to the scouringaction of beach material striking the structure; good strengthproperties, the ability to withstand sudden impact and deflec-tion under load; and availability in cross-section sizes of300 mm� 300 mm or greater and, often, lengths up to 15 m.
Historically, the combination of the above factors has resultedin the use of a short list of tropical hardwoods. Engineers havealmost always chosen a dense, naturally durable timber with aproven track record, such as greenheart or ekki, although afew other dense tropical hardwoods such as balau (Shoreaspp.), opepe (Nauclea diderrichii) and karri (Eucalyptyus diver-sicolor) have also been used. The disadvantage of this conser-vatism is that commercial exploitation of such a narrow rangeof timbers can accelerate their depletion and inflate the priceof certain tropical timbers (EA, 2010b).
Marketing LUS in a conservative market place has alwaystested the timber industry. End-user resistance to these LUS byengineers is being driven by a lack of (readily available) infor-mation about their technical properties. Two of the principalobstacles in using lesser-known species are that, either there islimited confidence in the pedigree of the technical informationabout these species, or little is known about their resistanceto marine borer attack (Williams et al., 2004b). It should beborne in mind that high natural durability in terrestrial con-ditions does not necessarily guarantee robust marine perform-ance (Cragg, 1996).
2. Hazards of the marine environmentTimber that is exposed in the marine environment below thehigh tide mark is subject to attack by marine bacteria, fungi
2
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
and wood-boring animals. Marine wood borers are a hetero-geneous group of wood-feeding organisms, including theTeredinidae (Distel, 2003) and Limnoriidae (Cragg, 2003).
The most economically important wood-boring Crustacea inEuropean waters belong to the Limnoriidae, isopods com-monly known as gribbles, which can cause significant damageto timber in the marine environment (Borges et al., 2014a,2014b; Castelló, 2011; Cragg, 2003). Wood-boring limnoriidshave evolved two key adaptations to use wood as substrate.The first is the ability to tunnel into timber for protection. Thesecond adaptation is to use timber as a food source (Kernet al., 2013).
Marine wood-boring Teredinidae (commonly known as ship-worms or teredinids) occur in almost all aquatic ecosystemsfrom tropical to cold-temperate waters (Turner, 1966). At leastnine wood-boring teredinids have been reported so far inEuropean coastal waters (Borges et al., 2014a).
Teredinids are estimated to cause over one billion dollars’worth of damage to submerged wooden structures per annum(Shipway et al., 2014) and the introduction of shipworms intonew areas is often followed by rapid and extensive destruction(Cohen and Carlton, 1995; Distel et al., 2011; Turner, 1966).The animals may cause severe damage to a timber structureover a comparatively short space of time.
Climate change has been causing a shift in the geographicalrange of species of marine borers (Parmesan and Yohe, 2003),but the effect of sea surface warming is still poorly understoodin marine systems (Rivadeneira and Fernandez, 2005).Therefore, the impact of global warming also needs to be con-sidered in relation to teredinid and limnoriid activity.
Increases in temperature and salinity have been reported inthe Mediterranean (Gibelin and Déqué, 2003; Sanchezand Gallardo, 2004). On the Portuguese coast, the increases insea surface temperature and salinity have been affecting theintertidal species range of marine borers (Borges et al., 2010).Increases in sea temperature and salinity are expected to con-tinue over the coming decades (Giannakopoulos and Le Sager,2009; Giorgi and Lionello, 2008), are known to extend teredi-nid distribution ranges (Borges et al., 2010, 2014a, 2014b;Paalvast and van der Velde, 2011; Shipway et al., 2014) andare likely to increase the vigour of limnoriids. Furthermore,warmer temperatures are known to accelerate growth and toincrease boring activity. Therefore, the risk posed by marineborers will become more severe due to global warming.
With reference to the coastal waters around the UK and basedon previous Trada research carried out in the 1960s (Hall andSaunders, 1967), there are two main groups of marine borers
and these are the teredinids and limnoriids. Limnoriids are ubi-quitous, whereas teredinids have previously been limited to thesouth coast and isolated estuarine areas along the west coast ofthe UK (Hall and Saunders, 1967). Plaster and Sawyer (1998)identified that changes in the species of marine borers andtheir distribution and occurrence in marine structures aroundthe UK coastline should be investigated due to improvingwater quality and predicted climate change. It should be notedthat since 1967 there has been no comprehensive survey of theUK coast for marine borer activity, and on the basis of evi-dence collated by Borges et al. (2014a, 2014b) changingenvironmental conditions do appear to affect marine borer dis-tribution and activity. The research cited above indicates thatclimate change may increase the risk of marine borers attack-ing timber structures around the UK coastline.
Moreover, anticipated increases in water temperature and estu-arine salinity (Paalvast and van der Velde, 2014) have thepotential to increase the vigour of these populations and toallow new species to invade. Furthermore, environmental legis-lation has led to a vast and continuing improvement in waterquality that has enabled marine borer populations to flourishin harbour areas where pollution had previously excluded them(Eaton and Hale, 1993). Therefore, it is reasonable to surmisethat the risks associated with marine borers are only going toincrease in the future.
The distribution of limnoriids (Figure 1) is controlled byenvironmental and biological factors as well as the presence ofwood. The most important environmental factors controlling
Figure 1. Specimens of Limnoria quadripunctata. Ventral view
3
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
the distribution and survival of limnoriids are temperature andsalinity (Borges et al., 2009; Eltringham, 1971). Temperatureand salinity influence the boring activity and the feeding ratesof limnoriids (Borges et al., 2009). Temperature is particularlyimportant during the reproductive and migratory season(Eltringham, 1971).
The life history strategy helps explain the distribution oflimnoriids. These organisms have low fecundity (Cragg, 2003)and tend to form large colonies in the more sheltered partsof timber structures. In temperate waters, swarming of theanimals tends to occur in the spring, although additionalswarming may occur in the autumn. They inhabit the surfaceof timber and they are fully mobile (Eaton and Hale, 1993)and tend to concentrate at the intertidal zone. They are verysmall animals, just about visible to the human eye whentimber is inspected in situ. Attack by limnoriids tends to besuperficial and results in the creation of an extensive networkof galleries at, or just below, the wood surface. In softwoods,the animals favour the less dense earlywood which can rapidlylead to the timber forming wafer-like plates (Eaton and Hale,1993).
Limnoriids tend to be sensitive to their environment and cancause significant damage in harbours and estuarine environ-ments characterised by waisting of the timber (Figure 2).However, where structures are exposed to the full force of thesea and where there is a high risk of mechanical abrasion, lim-noriids find it difficult to establish large populations. This isbecause the abrasive nature of the environment destroys thegalleries. In such instances, limnoriids tend to be restricted tothe sheltered parts of structures, particularly where joints areformed as these can provide suitable refuges for the animals(Harlow, personal communication, 2015).
Teredinids (Figure 3) are highly specialised bivalves charac-terised by elongated bodies and a greatly reduced shell, adap-tations that facilitate their wood-boring life style (Turner,1966). Unlike the limnoriids, the adults are sessile and theanimal remains in the same tunnel throughout its life. Eachtunnel is discrete and the animals avoid intruding into neigh-bouring tunnels as they grow and excavate into the timber.In warm waters some animals can grow in excess of 1 or 2 m(Eaton and Hale, 1993).
Even in timbers of limited volume, teredinids will not emergefrom the timber and will continue to bore alongside their neigh-bours until the timber is more or less destroyed and breaksapart (Eaton and Hale, 1993; Shipway et al., 2014). Teredinidsline their tunnels with a secretion of calcium carbonate. Theposterior part of the animal maintains contact with the exter-nal seawater environment through a fine hole generally <1 mmin diameter. This hole is the only external sign that shipwormshave colonised a timber component, which makes surveyingfor teredinids extremely difficult in situ. They are capable ofcausing extensive damage (Figure 4).
3. MaterialsEighteen LUS were selected for assessment on the basisof previous indicative laboratory and marine trials (Borgeset al., 2003; Crossman and Simm, 2004; Williams et al.,2004a), their commercial availability in suitable section sizes,volumes, price and evidence of legality and sustainability.The regions of origin of the candidate LUS, benchmark andwell-known timbers are presented in Tables 1 and 2 along with
Figure 2. Waisting of timber pile – that is, it develops a waist (or
a pencil point shape) in the intertidal zone, caused by limnoriid
attack. Note that there is no attack below the sea bed line which
is indicated by the arrow
Figure 3. Teredinid extracted from a Douglas fir waling of the
Barmouth viaduct within the intertidal zone of the estuary mouth
of the River Mawddach, Gwynedd
4
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
the reference codes used to identify the timbers throughoutmarine exposure trials.
Ekki and greenheart are emboldened to highlight their impor-tance as the benchmarking timbers. The sapwood of Europeanredwood (Pinus sylvestris), also emboldened, was used forcontrol purposes to confirm the vigour of the marine borersduring the laboratory and marine trials and also to validatethe experimental data. European oak (Quercus spp.) was
included in the project on the basis that it is a commerciallyimportant, naturally durable European hardwood. Other trop-ical timbers, well known for their use in the marine environ-ment and/or their natural durability, listed in Table 2 wereselected for comparative purposes.
Stock material was selected at random by both suppliers wherea reasonable attempt was made to ensure that the selectedboards were widely dispersed within a timber parcel or parcelsso that the natural variability of timber within commercialsupplies was reproduced. Ten boards comprising stock materialof �25 mm� 125 mm� 500 mm dimensions were provided bythe timber suppliers of each timber species. One small speci-men was cut from one of the boards of each species so that thespecies/genus could be confirmed using microscopy techniques.
It should be noted that in almost all cases, identification downto the genus level was possible. However, it was not practicableto identify individual species as most commercial timbers com-prise groups of species. However, on the basis of an examin-ation of the microscopical properties and a consideration ofthe physical properties of the specimens, all candidatesmatched their reported commercial names.
4. MethodsFast-track novel laboratory screening trials to determinecomparative marine borer and abrasion resistance have beendevised by the University of Portsmouth and the Forest Products
Figure 4. Example of the typical damage caused by teredinids
Commercial name Botanical name Region of origin Supplier Reference code
Angelim vermelho Dinizia excelsa South America EcoChoice Ltd AVBasrolocus Dicoryia guianensis South America EcoChoice Ltd BACloeziana Eucalyptus cloeziana South Africa EcoChoice Ltd CLCupiuba Goupia glabra South America EcoChoice Ltd CUDabema (dahoma) Piptadeniastrum africanum West Africa EcoChoice Ltd DAEvuess (kruma) Klainidoxa gabonensis West Africa EcoChoice Ltd EVGarapa Apuleia leiocarpa South America EcoChoice Ltd GAMassaranduba Manilkara spp. South America EcoChoice Ltd MAMora Mora excelsa South America Aitken and Howard Ltd MOMukulungu Autronella congoensis West Africa EcoChoice Ltd MUNiove Staudtia kamerunensis West Africa EcoChoice Ltd NIOkan (denya) Cylicodiscus gabunensis West Africa EcoChoice Ltd OKPiquia Caryocar spp. South America EcoChoice Ltd PISapucaia Lecythis paraensis South America EcoChoice Ltd SASouge Parinari excelsa West Africa EcoChoice Ltd SOTali Erythrophleum ivorense West Africa EcoChoice Ltd TATatajuba Bagassa spp. South America EcoChoice Ltd TJTimborana Enterolobium schomburgkii South America EcoChoice Ltd TI
Table 1. List of less-used timber species selected for evaluation
5
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
Research Centre (Borges et al., 2008; Sawyer and Williams,2005) to, respectively, assess the feasibility of determining lim-noriid and abrasion resistance. However, marine exposure trialsare still required to determine comparative resistance to teredi-nid attack due to the difficulty of maintaining viable teredinidpopulations under laboratory conditions. The rationale behindboth laboratory trials was ‘if it fails in the laboratory then it islikely to fail in the sea’ and that these trials could filter outpoor performing timbers before progressing to more expensive,time-consuming marine exposure trials.
4.1 Investigating limnoriid resistanceThe faecal pellet production rate of gribble matches the feedingor ingestion rate quite closely and it is much easier to measurethan the wood loss by ingestion. The approach used in thisinvestigation takes advantage of this situation. The methoddescribed in this report has been developed from investigationsof the feeding biology of these organisms (Wykes et al., 1997)and the evaluation of the test conditions for this method(Borges et al., 2009; Praël et al., 1999) and implemented inprevious research undertaken by Borges et al. (2003, 2008).Their research demonstrated that determining limnoriid resist-ance in fast-track laboratory trials was reliable and the methodused follows that of these authors closely.
Specimens of Limnoria quadripunctata Holthuis were obtainedfrom a laboratory population that was maintained in blocks ofEuropean redwood kept in running seawater from LangstoneHarbour near Portsmouth, UK. Animals were transferred to acell culture dish using a fine sable brush or fine forceps.
Trial specimens were sticks measuring 20 mm� 4·5 mm�2 mm prepared from the stock material. Heartwood was usedin all cases except for the control timber (European redwood),
for which sapwood was used. The vigour and health of theanimals were assessed and confirmed by observing feedingrates on the sapwood blocks prior to the starting the trials.Confirmation of high feeding rates on sticks of Europeanredwood sapwood validated the experiment.
Prior to experimentation, the sticks were leached in seawaterfor 1 week, with a change of water after 3 d to remove water-soluble extractives that were not relevant to long-term perform-ance of naturally durable timbers. Each stick was placed into4 ml of seawater in each chamber of a cell culture box contain-ing 12 chambers measuring 20 mm in diameter. The cellculture dishes were kept in the laboratory under ambient light-ing conditions at 20± 2°C for 28 d.
One animal was placed into each well of each dish. Afterhalf a week, animals and sticks were carefully transferred tomatching cell culture dishes with fresh seawater and thenumber of faecal pellets left in the original dishes was countedusing image analysis software (ImageJ, 2016) to analyse digitalmacro images of the chambers viewed from above and illumi-nated from below using a stereo microscope. Figures 5 and 6illustrate the typically low and high feeding rates that wereobserved.
The rationale behind this laboratory test is that the faecalpellet production rate matches the feeding or ingestion ratesquite closely. A one-way analysis of variance (Anova) was per-formed with Dunnet’s post-hoc test to identify timbers forwhich feeding rates were significantly
& lower than on ekki& lower than on greenheart& higher than on ekki& higher than on greenheart.
Commercial name Botanical name Region of origin Supplier Reference code
Yellow Balau Shorea spp. S E Asia (Sabah) Aitken and Howard Ltd BUDouglas fir Pseudotsuga menziessii North America Aitken and Howard Ltd DFBilinga Nauclea diderrichii West Africa EcoChoice Ltd BIEkki Lophira alata West Africa Aitken and Howard Ltd EGreenheart Chlorocardium rodiei Guyana Aitken and Howard Ltd GHKarri Eucalyptus diversicolor Australia Aitken and Howard Ltd KAOak Quercus spp. Europe Aitken and Howard Ltd ALOpepe Nauclea diderrichii West Africa Aitken and Howard Ltd OPPurpleheart Peltogyne spp. South America Aitken and Howard Ltd PUEuropean redwood Pinus sylvestris Western Europe Trada technology SP
Table 2. List of timber species with a known track record for use
in marine construction including benchmark and control species
6
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
The results were presented as average daily feeding rates andthese were compared against the results obtained for the twobenchmark timbers, greenheart and ekki.
4.2 Investigating abrasion resistanceTest blocks were prepared from the timber species listed inTables 1 and 2 for testing by immersion in running seawaterfor 4 weeks to become saturated. Moisture meter checks with acalibrated electrical resistance type moisture meter indicatedthat all blocks were above the fibre saturation point with read-ings in test blocks in excess of 30%. The fibre saturation point
is a theoretical state where the wood cell walls are fully satu-rated with water and the cavities are empty. Typically, thisvalue is �27% (Desch and Dinwoodie, 1996). The fibre satur-ation point can be influenced by density. The fibre saturationpoint deceases with increasing density (Skaar, 1988).Immediately prior to testing, the volume of the blocks wasmeasured by displacement in a eureka can. Fresh water dis-placed by the blocks was weighed directly.
A total of six test specimens per species were prepared. Eachblock was drilled centrally with an 8 mm hole for securing tothe test frames as illustrated in Figure 7. The testing machineused was adapted from a Los Angeles aggregate fragmentationresistance apparatus. This comprised a heavy robust steel drumthat can be sealed so as to be watertight. The speed of rotationhad been determined from previous research (Sawyer andWilliams, 2005) and was 33 r/min. The testing regime used aninitial charge of 25 kg of 20 mm flint shingle with 15 litres ofseawater to ensure a consistent abrasive environment.
After 80 000 revolutions, the shingle was emptied andthoroughly flushed. Fresh shingle and seawater were loadedand the machine run for a further 80 000 revolutions. In thisway, each test had a duration of 5 d. At the completion oftesting, all blocks were thoroughly washed and cleaned andstored under water to prevent drying. Finally, the volume lossof blocks was determined by the displacement method.Figure 8 illustrates typical erosion of the specimens.
The arrangement of the LUS test blocks on the racksfollowed a Latin square distribution so that each timber wasreplicated at all the rack positions. The variation in volume
Figure 5. Low feeding rate observed in a chamber containing a
stick of ekki and a limnoriid. In this case, ingestion of wood has
killed the limnoriid
Figure 6. High feeding rate observed in a chamber containing a
stick of European redwood sapwood which confirmed the vigour
of the limnoriid population
Figure 7. Typical test frame illustrating the arrangement of test
blocks prior to the start of the experiment
7
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
loss between the different candidate timbers was examinedwith a general linear model Anova, with the candidatetimber and the position on the rack as fixed factors. Dunnet’spost-hoc test was carried out to identify timbers that per-formed significantly
& better than ekki& better than greenheart& worse than ekki& worse than greenheart.
4.3 Investigating teredinid resistanceThe test site for the marine exposure trial was located in Olhãoharbour on the Ria Formosa lagoon, Portugal. The conditionsin the Ria Formosa lagoon fall well within the environmentallimits described by Borges et al. (2014a) that will supportmarine boring organisms.
The tidal regime at this location is semi-diurnal with a rangeof 1·35 m on neap tides to 3 m on spring tides. Although thelagoon is on the Atlantic coast, the climatic conditions areessentially Mediterranean with hot dry summers and warm wetwinters. The average temperatures and salinities of the lagoonwater at Olhão (Newton and Mudge, 2003) range between12 and 28°C, and between 33 and 36·5 psu, respectively.Previous research by Williams et al. (2004a) reported the pres-ence of aggressive teredinid and limnoriid attack on vulnerabletimbers.
Six specimens of each timber were cut to dimensions of20 mm� 75 mm� 300 mm and were prepared for immersion.European redwood sapwood was used for control purposes and
to assess the vigour of borer activity. BS EN 275 (BSI, 1992)requires that samples are arranged vertically. The test method inthis trial was an improvement on BS EN 275 (BSI, 1992) as theracks were immersed horizontally, which ensured that allsamples were exposed to the same tidal marine conditions, withthe samples orientated vertically. The test racks were weightedand suspended �0·5 m from the sea bed.
The marine exposure trial lasted for 18 months, from March2008 to September 2009. Three assessment visits were made tomonitor performance. During each assessment visit, the racksand samples were cleaned of all marine fouling. This was toensure that calcareous surface growths could not influenceinterpretation of the X-ray photographs. After cleaning, thespecimens were examined for signs of attack by marine borers.Figures 9 and 10 illustrate examples of test racks before andafter cleaning.
The racks were then wrapped in polyethylene bags to preventstress to teredinid populations in the timber samples caused by
Figure 8. A test frame holding eroded greenheart test blocks
after 160 000 revolutions
Figure 9. Typical marine fouling on a test rack after 6 months’
immersion
8
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
drying out of the timber during transportation between thetest site and a local X-ray clinic. The duration betweenremoval from the site and re-immersion following X-ray pho-tography was no more than 24 h. Teredinid attack was assessedusing the visual assessment categories detailed in Table 3,
based on the methodology given in BS EN 275 (BSI, 1992).The whole specimen was assessed, with the exception of thetimber coincident with the frames that absorbed the X-rays.The exposure variables were determined by the equipmentused, but images were a good indication of the desired result,so X-ray exposure was determined by trial and error.
For each timber, the visual assessment ratings for each testsample were added and the result divided by the number oftest samples to yield a notional average rating. As results forthe marine exposure trial were reported in terms of an arbi-trary scale (a practical and widely used approach for field trialswith timber), a parametric statistical approach is not ap-propriate, so standard error values were not quoted.
5. Results
5.1 Resistance to limnoriidsThe data in Figure 11 show pellet production rate per dayover a 28 d period (mean±standard error). Significantly lowerfeeding rates than those recorded on both benchmark timbers(greenheart and ekki) were recorded on niove (Staudtia kamer-unensis), balau and cupuiba (Goupia glabra). Significantlyhigher feeding rates than both benchmark timbers wererecorded on purpleheart (Peltogyne spp.), mukulungu(Autronella congoensis), eveuss (Klanidoxa gabonensis) andEuropean redwood (Scots pine/Pinus sylvestris). Significantlyhigher rates than that recorded for greenheart alone wererecorded on European oak, tali (Erythrophleum ivorense) andangelim vermelho (Dinizia excelsa).
5.2 Resistance to abrasionThe data in Figure 12 show the average loss in volume ofthe timber species. Comparison of abrasion data is presentedas a measure of mean volume loss (±standard error) of thetimbers against the mean volume loss measured for ekki andgreenheart. With reference to ekki, none of the candidatetimbers indicated that they performed significantly better and
Figure 10. General appearance of a test rack after cleaning to
facilitate X-ray photography
Numericalassessment category Amount of attack caused by teredinid as percentage board volume
0 No attack1 Minor attack. Single or a few scattered tunnels not covering more than 10% of the specimen areas as it
appears on X-ray film2 Moderate attack. Tunnels not covering more than 25% of the specimen area as it appears on X-ray film3 Severe attack. Tunnels covering between 25% and 50% of the area of the specimen as it appears on
X-ray film4 Failure. Tunnels covering more than 50% of the area of the specimen as it appears on X-ray film
18 out of 26 species (greenheart to balau on Figure 12) per-formed significantly worse than ekki when challenged withshingle. With reference to greenheart, souge (Parinari excels),oak, eveuss and ekki indicated that they performed signifi-cantly better during this trial. Only cupiuba, opepe, massaran-duba (Manilkara spp.), bilinga (Nauclea diderrichii) and yellowbalau performed significantly worse than greenheart.
5.3 Resistance to teredinidsThe data in Figure 13 show the mean visual assessment ratingsrecorded on all timber species exposed during the 18-monthmarine trial period. The failure of the sapwood controlblocks demonstrated that the test site has aggressive teredinidpopulations. Furthermore, oak, Douglas fir (Pseudotsugamenziessii), karri, mora (Mora excelsa) and purpleheartalso yielded a maximum mean visual assessment ratingof 4·0.
6. Discussion
6.1 Resistance to limnoriidsThe sapwood of European redwood was used as a control tovalidate the experiment. The comparatively high daily feedingrates observed on sapwood confirmed the vigour of the testorganisms and validated this laboratory trial.
A comparison of timber performance, expressed as suppressionof feeding rates, against the benchmark species greenheart andekki, provides an indication of their comparative resistance to
limnoriid attack. From a benchmarking point of view it can beseen that if significant differences in performance to greenheartare chosen as a selection threshold, then a comparatively highnumber of candidate species would be rejected at this stage.However, if significant differences in performance to ekki areused as the selection threshold, then only purpleheart, muku-lungu and evuess would be considered for rejection at thescreening stage. Given that ekki is one of the favoured timbersused for marine and freshwater construction, it would bereasonable to use ekki as the benchmarking species on thebasis of these results alone.
The one-way Anova identified a number of LUS of timber asperforming significantly better than ekki and greenheart. Itshould be borne in mind that good resistance to limnoriidsdoes not necessarily mean that the timbers will have goodresistance to attack by teredinids.
6.2 Resistance to abrasionThe sapwood of European redwood was not included in theabrasion trials as a control as there was no requirement to vali-date the vigour of test organisms in this instance. However,Douglas fir was included as a reference timber because it is acommercially important softwood that is used for marine con-struction. Furthermore, there is anecdotal evidence to suggestthat Douglas fir performs better than some denser, harder tro-pical hardwoods as the less dense softwood structure can actas a shock absorber that more efficiently dissipates the energyexpended on the timber surface under shingle impact.
0
0·5
1·0
1·5
2·0
Mea
n vi
sual
ass
essm
ent
ratin
g
2·5
3·0
3·5
4·0O
ak
Ang
elim
ver
mel
ho
Basr
aloc
us
Bilin
ga
Yello
w b
alau
Clo
ezia
na
Cup
iuba
Dab
ema
Dou
glas
fir
Ekki
Eveu
ss
Gar
apa
Gre
enhe
art
Kar
ri
Mas
sara
duba
Mor
a
Muk
ulun
gu
Nio
ve
Oka
n
Ope
pe
Euro
pean
red
woo
d
Piqu
ia
Purp
lehe
art
Sapu
caia
Soug
e
Tali
Tim
bora
na
Tata
juba
Figure 13. Mean visual assessment ratings recorded
after18 months exposure
11
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
The effects of abrasion in marine trials are difficult to studydue to the extreme variability of weather and site conditions,and existing laboratory methods also bear little resemblanceto the effects of marine abrasion. In relation to abrasion, itshould be noted that Oliver and Woods (1959) undertook aninvestigation to determine the rates of wear by shingle abrasionof a number of timber species used as piles and plankingin sea defence groynes. Their study spanned a 6-year period(1953–1959). Again, projects of this duration are too longto be commercially practicable. Other than the work of Oliverand Woods (1959), a previous search of the literature bySawyer and Williams (2005) found that there had been littleother research on the resistance of marine timbers to theeffects of marine abrasion. The abrasion tests exposed thetimbers to 160 000 revolutions. If it is assumed that timber insitu is exposed to six waves every minute, it follows that thestructure will be exposed to 360 waves per hour.
On the basis of these assumptions and in the absence of otherinformation, the test of 160 000 cycles would therefore equateto about 450 tides. It is unlikely that all waves will have suffi-cient energy to subject the structure to abrasion with everywave cycle. Therefore, a second assumption has to be madethat the waves will only generate abrasive conditions for 33%of the time. This would then equate to 1350 tides or about675 d. The abrasion tests described in this report have there-fore condensed 675 d in service (almost 2 years) into a 5 d testperiod. However, it should be borne in mind that direct com-parisons to exposure in situ are difficult to make and theassumptions detailed above could be viewed as indicative andprovide a means to predict comparative abrasion resistance ofLUS used in groyne structures in the intertidal zone on ashingle beach.
From a benchmarking point of view, it can be seen that if asignificant difference in performance compared with green-heart is chosen as a selection threshold then three LUS timbersperformed significantly better than greenheart. These were tali,eveuss and souge. However, if ekki is chosen as a selectionthreshold, all candidate species would be rejected at this stageas none performed significantly better than ekki if abrasionis the principal material property required by the designengineer.
It is possible that different results could be obtained if the can-didate timbers were exposed to sand abrasion. However,shingle was selected on the basis that it provided the worst-casescenario.
6.3 Resistance to teredinidsStatistical analysis using Anova was not possible with thedata obtained from the marine exposure trial as a non-linearvisual assessment scale has been used. The visual assessment
rating scale has been developed to identify shipworm resistancetowards the lower end of the scale so that differences betweenratings of 1·0 and 2·0 are easily distinguishable in the field.The visual assessment scheme described in Table 3 is skewedtowards the lower end of the assessment scale. This is becauseit is important to separate those samples with <10% colonisa-tion and those with 10–25% colonisation. It is this area wherethe interpretation of results may provide the basis for differen-tiating timber species that may be classified as being resistantto teredinid attack (<10% colonisation) and moderately resist-ant (10–25% colonisation).
Greater than 25% colonisation indicates severe attack. There islittle point in defining differences above this threshold level asthe timber species with a rating above this level may be con-sidered as not having any practical resistance to teredinidattack.
The effective use of X-radiography as a method of assessmentrelies on thorough cleaning of the timber samples as the rem-nants of calcareous material can lead to misinterpretation withthe early stages of colonisation by teredinids. As the trials pro-gress, the juvenile teredinids grow and extend their tunnels.As they grow, they become more distinctive when assessedusing X-radiography. As a consequence, small variations in themean visual assessment ratings of candidate timber during theearly stages of the trial may occur due to juvenile mortality.Figure 14 illustrates severe teredinid attack clearly demon-strated by extensive tunnelling and Figure 15 illustrates good
Figure 14. Extensive teredinid attack in a control specimen
of European redwood detected by X-ray photography. The
extensive damage confirms the presence of an aggressive
teredinid population in the Ria Formosa lagoon and validates
the marine trial
12
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
resistance to shipworm attack as demonstrated by the lack oftunnels in the specimen.
After 18 months of exposure in a hazardous marine environ-ment, the following candidate timbers performed comparablyto ekki and greenheart, in that no shipworm attack wasdetected in them: angelim vermelho, basralocus (Dicoyriaguianensis), okan (Cyclodiscus gabunensis) and tali.
Those timbers exhibiting minor attack by shipworm werecupiuba, eveuss, garapa (Apuleia leiocarpa), massaranduba(Manilkara spp.), piquia (Caryocar spp.), sapucaia (Lecythisparaensis), souge and timborana (Enterolobium schomburgkii).However, after 18 months of exposure, the mean visual assess-ment ratings of these timbers were still <1·0, which indicatedthat these candidate timbers may be considered as resistant toattack by teredinids, despite performing slightly worse thangreenheart and ekki.
Only dabema (Piptadeniastrum africanum) and mukulunguwere considered as being moderately resistant to teredinidattack. In this particular trial, those timber species that yieldeda mean visual assessment rating in excess of 2·0 were con-sidered to have little resistance to attack by teredinids. Thetimbers that fall into this category are cloeziana (Eucalyptuscloeziana), mora, niove and tatajuba (Bagassa spp.).
7. ConclusionsMarketing LUS has always tested the timber industry andthere has often been considerable end-user resistance toworking with LUS as their technical properties are not fullyappreciated. The data and conclusions contained within this
paper should begin to erode this conservatism and encouragethe use of a wider range of LUS in marine applications, andcontribute towards the goal of profitable forestry and sustainedtimber yield management. If value can be added to a tropicalforest for timber production then it may come under lesspressure for conversion to other uses.
However, the acceptance of LUS will require a holisticapproach where, in addition to considering marine borer resist-ance and abrasion resistance, other factors such as strength,price, section sizes, shrinkage, movement in service, workabilityand machinability characteristics, and delivery times may influ-ence choice.
The selection of timber by understanding service require-ments and hazards may provide a means to introduce LUSinto marine construction. In other words, the functional per-formance of a timber and its ability to withstand the mostdominant site-specific hazards, whether it be resistance to lim-noriids, teredinids or abrasion, could drive the selection ofLUS.
Selection by this means could be undertaken with thefull support of an examination and assessment regimewhereby the performance of these timbers can be monitoredover their service life so reliable data from an asset manage-ment point of view may be collated. Of course, this requires along-term vision. Three LUS – tali, souge and evuess – havebeen installed on beach groynes at Pevensey Bay in 2009, aspart of an Environment Agency project. The performance ofthese species was evaluated in Spring 2016.
The screening programmes described in this paper providean effective mechanism for the rapid evaluation of both themarine borer and abrasion resistance of LUS. However, thescreening test described in this paper does not solve the signifi-cant problem of determining the strength properties of LUS.Confidence in strength properties is a critical factor that engin-eers require when specifying timber, and a dearth of reliableinformation regarding the strength of LUS is one of the majorobstacles in preventing their use. Strength test programmesrequire significant investment.
Historically, there has been considerable end-user resistanceto using LUS as their technical properties are not fully ap-preciated. The conclusions that identify the technical prop-erties of a range of LUS presented in this paper may beginto encourage the use of a wider range of less-used timbers.The research also provides robust evidence to justify investmentin testing programmes to determine the characteristic valuesfor bending strength, modulus of elasticity and density usingthe requirements of BS EN 384 (BSI, 2004) and BS EN 408(BSI, 2003).
Figure 15. No evidence of teredinid attack in a specimen of the
LUS angelim vermelho
13
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
Borges LMS, Cragg SM and Williams JR (2003) Comparing theResistance of a Number of Lesser Known Species ofTropical Hardwoods to the Marine Borer LimnoriaUsing a Short Term Laboratory Assay. InternationalResearch Group on Wood Preservation, Stockholm,Sweden, Document No. IRG/WP 03-10500.
Borges LMS, Cragg SM, Bergot J et al. (2008) Laboratoryscreening of tropical hardwoods for natural resistanceto the marine borer Limnoria quadripunctata with aninvestigation of the role of leachable and non-leachablefactors. Holzforschung 62(1): 99–111.
Borges LMS, Cragg SM and Busch S (2009) A laboratoryassay for measuring feeding and mortality of themarine wood-borer Limnoria under forced feedingconditions: a basis for a standard test method.International Biodeterioration and Biodegradation 63(3):289–296.
Borges LMS, Valente AA, Palma P and Nunes L (2010) Changesin the wood boring community in the Tagus Estuary:a case study. Marine Biodiversity Records 3: e41.
Borges LMS, Merckelbach LM, Sampaio I and Cragg SM (2014a)Diversity, environmental requirements and biogeographyof bivalve wood-borers (Teredinidae) in European coastalwaters. Frontiers in Zoology 11: 13.
Borges LMS, Mercklebach L and Cragg SM (2014b)Biogeography of wood-boring crustaceans (Isopoda:Limnoriidae) established in European coastal waters.PLoS ONE 9(10): e109593.
Brazier A (1995) Greenheart and the Choice of Timbers forMarine Work: Half Day Meeting at the Institution ofCivil Engineers (Maritime Board). Thomas Telford,London, UK, http://dx.doi.org/10.1680/thitmeacftcu.45422.0003.
BSI (1992) BS EN 275:1992: Wood preservatives.Determination of the protective effectiveness againstmarine borers. BSI, London, UK.
BSI (2003) BS EN 408:2003: Timber structures – structuraltimber and glued laminated timber – determinationof some physical and mechanical properties. BSI,London, UK.
BSI (2004) BS EN 384:2004: Structural timber – determinationof characteristic values of mechanical properties anddensity. BSI, London, UK.
Castelló J (2011) The genus Limnoria (Limnoriidae, Isopoda,Crustacea) in Europe, including a key to species. Zootaxa2968: 1–25.
Cohen AN and Carlton JT (1995) Biological Study:Non-Indigenous Aquatic Species in a United StatesEstuary: A Case Study of the Biological Invasions of theSan Francisco Bay and Delta. University of California,National Technical Information Service, Springfield, VA,USA, Final Report.
Cragg SM (1996) Timber in the marine environment. TimberTrades Journal 376: 26–28.
Cragg SM (2003) Marine wood boring arthropods: ecology,functional anatomy and control measures. In WoodDeterioration and Preservation: Advances in our ChangingWorld (Goodell TP, Nicholas B and Schultz DD (eds)).American Chemical Society, Washington, DC, USA,pp. 272–288.
Crossman M and Simm J (2004) Manual on the Use of Timberfor Coastal and River Engineering. HRWallingford,Thomas Telford, London.
Desch HE and Dinwoodie JM (1996) Timber, Structure,Properties, Conversion and Use, 7th edn. Macmillan PressLimited, London, UK.
Distel DL (2003) The biology of marine wood boring bivalvesand their bacteria endosymbionts. In Wood Deteriorationand Preservation: Advances in our Changing World(Goodel TP, Nicholas B and Schultz DD (eds)).American Chemical Society, Washington, DC, USA,pp. 253–271.
Distel DL, Amin M, Burgoyne A et al. (2011) Molecularphylogeny of Pholadoidea Lamarck, 1809 supports a singleorigin for xylotrophy (wood feeding) and xylotrophicbacterial endosymbiosis in Bivalvia. MolecularPhylogenetics and Evolution 61(2): 245–254.
EA (Environment Agency) (2010a) Alternative HardwoodTimbers for Use in Marine and Fresh Water Construction.(Williams JR, Meaden M and Simm J (eds)).Environment Agency, Bristol, UK, Project: SC070083/R1.
EA (2010b) The Coastal Handbook: A Guide for AllThose Working on the Coast. Environment Agency,Bristol, UK.
Eaton RA and Hale MDC (1993) Wood: Decay Pests andProtection. Chapman and Hall, New York, NY, USA.
Eltringham SK (1971) Factors affecting the distribution ofthe burrows of the marine wood-boring Isopod Limnoria.International Biodeterioration 7(2): 61–67.
Giannakopoulos C and Le Sager P (2009) Climatic changes andassociated impacts in the Mediterranean resulting from a2°C degrees global warming. Global Planet Change 68(3):209–224.
Gibelin AL and Déqué M (2003) Anthropogenic climate changeover the Mediterranean region simulated by a globalvariable resolution model. Climate Dynamics 20(4):327–339.
Giorgi F and Lionello P (2008) Climate change projections forthe Mediterranean region. Global Planet Change 63(2–3):90–104.
Hall GS and Saunders RG (1967) Incidence of Marine BorersAround Britain’s Coasts. Trada Research, High Wycombe,UK, Report B/RR/4.
ImageJ (2016) An open platform for scientific image analysis.See http://imagej.net/downloads (accessed 23/09/2016).
14
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
Kern M, McGeehan JE, Streeter SD et al. (2013) Structuralcharacterization of a unique marine animal family 7cellobiohydrolase suggests a mechanism of cellulase salttolerance. Proceedings of the National Academy of Sciencesof the United States 110(25): 10189–10194.
Newton A and Mudge SM (2003) Temperature and salinityregimes in a shallow, mesotidal lagoon, the Ria Formosa,Portugal. Estuarine, Shelf and Coastal Science 56(1–2):73–85.
Oliver AC and Woods RP (1959) The Resistance of CertainTimbers in Sea Defence Groynes to Shingle Abrasion.The Timber Development Association (TDA),High Wycombe, UK, Test Record B/TR/4.
Paalvast P and van der Velde G (2011) New threats of an oldenemy: the distribution of the shipworm Teredo navalis L.(Bivalvia: Teredinidae) related to climate change in thePort of Rotterdam area, the Netherlands. Marine PollutionBulletin 62(8): 1822–1829.
Paalvast P and van der Velde G (2014) Long termanthropogenic changes and ecosystem service consequencesin the northern part of the complex Rhine-Meuse estuarinesystem. Ocean and Coastal Management 92: 50–64.
Parmesan C and Yohe G (2003) A globally coherent fingerprintof climate change impacts across natural systems. Nature421(6918): 37–42.
Plaster SEM and Sawyer GS (1998) A survey to assess the currentand future usage of timber in British port structures.Journal of the Institute of Wood Science 14(6): 287–292.
Praël A, Cragg SM and Eaton RA (1999) Development of a NovelApproach to Testing of Wood Preservatives for Marine Use:Effects of Temperature on Feeding Rates of Limnoria.International Research Group on Wood Preservation,Stockholm, Sweden, IRG/WP/99-10325.
Rivadeneira MM and Fernandez M (2005) Shifts in southernend points of distribution in rocky intertidal species alongthe south-eastern Pacific coast. Journal of Biogeography32(2): 203–209.
Sanchez E and Gallardo C (2004) Future climate extreme eventsin the Mediterranean simulated by a regional climatemodel: a first approach. Global and Planet Change 44(1–4):163–180.
Sawyer GS and Williams JR (2005) An Investigation to Assessthe Feasibility of Developing an Accelerated LaboratoryTest to Determine the Abrasion Resistance of Lesser-UsedTimber Species for Use in Marine Construction.International Research Group on Wood Preservation,Stockholm, Sweden, IRG/WP 05- 20317.
Shipway JR, Borges LMS, Muller J and Cragg SM (2014)The broadcast spawning Caribbean shipworm, Teredothyradominicensis (Bivalvia, Teredinidae), has invaded andbecome established in the eastern Mediterranean Sea.Biological Invasions 16(10): 2037–2048.
Skaar C (1988) Wood–Water Relations. Springer Verlag,Berlin, Germany.
Turner RD (1966) A Survey and Illustrated Catalogue of theTeredinidae. Museum of Comparative Zoology,Harvard University, Cambridge, MA, USA.
Williams JR, Sawyer GS, Cragg SM and Simm J (2004a) AQuestionnaire Survey to Establish the Perceptions of UKSpecifiers Concerning the Key Material Attributes ofTimber for Use in Marine and Fresh Water Construction.International Research Group on Wood Preservation,Stockholm, Sweden, IRG/WP 04-10519.
Williams JR, Cragg SM, Borges LMS and Icely JD (2004b) MarineExposure Assessment of the Natural Resistance of aNumber of Lesser Known Species of Tropical Hardwoods toTeredinid and Limnoriid Borers. International ResearchGroup on Wood Preservation, Stockholm, Sweden,IRG/WP 04-10520.
Wykes C, Cragg SM and Pitman AJ (1997) The FunctionalAnatomy of the Digestive Caecae and Gut Residence Timesfor the Marine Borer Limnoria (Crustacea: Isopoda).International Research Group on Wood Preservation,Stockholm, Sweden, IRG/WP/97-10204.
HOW CAN YOU CONTRIBUTE?
To discuss this paper, please email up to 500 words to theeditor at [email protected]. Your contribution will beforwarded to the author(s) for a reply and, if consideredappropriate by the editorial board, it will be published asdiscussion in a future issue of the journal.
Proceedings journals rely entirely on contributions fromthe civil engineering profession (and allied disciplines).Information about how to submit your paper onlineis available at www.icevirtuallibrary.com/page/authors,where you will also find detailed author guidelines.
15
Construction Materials Evaluating less-used timber speciesfor marine constructionWilliams, Sawyer, Cragg et al.
![Comparison of Extraction Methods of Chitin from Ganoderma ... · 2 BioMedResearchInternational inbioreactorswithallautomatedandcontrolledconditions; therefore,mycelialbiomassproducedineachbatchishomo-geneous,inqualityandquantity[6,12].](https://static.documents.pub/doc/80x56/5fc907d014ef0f43a247f007/comparison-of-extraction-methods-of-chitin-from-ganoderma-2-biomedresearchinternational.jpg)
