See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326992930

Performance of CPT-based methods to assess monopile driveability in North

Sea sands

Article in Ocean Engineering · August 2018

DOI: 10.1016/j.oceaneng.2018.08.010

CITATIONS READS

18 1,676

6 authors, including:

Kenneth Gavin Luke J. Prendergast

Delft University of Technology University of Nottingham
223 PUBLICATIONS 5,185 CITATIONS 83 PUBLICATIONS 1,534 CITATIONS

SEE PROFILE SEE PROFILE

Paulo Cachim Paul Doherty

University of Aveiro University College Dublin
91 PUBLICATIONS 2,156 CITATIONS 44 PUBLICATIONS 1,501 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Luke J. Prendergast on 13 August 2018.

The user has requested enhancement of the downloaded file.

 Published in Ocean Engineering 166 2018 pp.76-91

1 Performance of CPT-based methods to assess monopile driveability in North Sea sands

2 Byrne, T.a,1, Gavin, K.b,2, Prendergast, L.J.b,c,3,*, Cachim, P.d,e,4, Doherty, P.f,5, Chenicheri Pulukul, S.g,6
a
3 School of Civil Engineering,
4 University College Dublin,
5 Belfield,
6 Dublin 4,
7 Ireland
8
b
9 Faculty of Civil Engineering and Geosciences,
10 Delft University of Technology,
11 Building 23,
12 Stevinweg 1 / PO-box 5048,
13 2628 CN Delft / 2600 GA Delft,
14 The Netherlands
15
c
16 Department of Civil Engineering,
17 Faculty of Engineering,
18 University of Nottingham,
19 Nottingham,
20 NG7 2RD,
21 United Kingdom
22
d
23 RISCO & Department of Civil Engineering,
24 University of Aveiro,
25 3810-193 Aveiro,
26 Portugal
27
e
28 Laboratory of Geotechnics,
29 Department of Civil Engineering,
30 University of Ghent,
31 Belgium
32
f
33 Gavin and Doherty Geosolutions,
34 Unit A2,
35 Nutgrove Office Park,
36 Rathfarnham,
37 Dublin 14,
38 Ireland
39
g
40 Shell UK Ltd,
41 1 Altens Farm Road,
42 Nigg,
43 Aberdeen,
44 AB12 3FY,
45 United Kingdom
46
*
47 Corresponding author
48
49 Email: [email protected], [email protected], [email protected],
4
50 [email protected], [email protected], [email protected]
51
52
 Published in Ocean Engineering 166 2018 pp.76-91

53 Abstract

54 Offshore pile driving is a high-risk activity as delays can be financially punitive. Experience of pile
55 driving for offshore jacket structures where pile diameters are typically < 2m has led to the development
56 of empirical pile driveability models with proven predictive capability. The application of these
57 methods to larger diameter piles is uncertain. A major component of driveability models involves
58 estimating the static resistance to driving, SRD, a parameter analogous to pile axial capacity. Recent
59 research on axial capacity design has led to improved models that use Cone Penetration Test, CPT data
60 to estimate pile capacity and include for the effects of friction fatigue and soil plugging. The
61 applicability of these methods to estimating pile driveability for larger diameter piles is of interest. In
62 this paper, recent CPT based axial capacity approaches, modified for mobilised base resistance and
63 ageing, are applied to estimating driveability of 4.2m diameter piles. A database of pile installation
64 records from North sea installations are used to benchmark the methods. Accounting for factors such
65 as pile ageing and the relatively low displacement mobilised during individual hammer blows improves
66 the quality of prediction of pile driveability for the conditions evaluated in this study.

67

68 Keywords: Pile Driveability; Static Capacity; UWA-05; IC-05; Sand; Monopiles; Offshore Wind; Base
69 Resistance-Settlement

70

71 1. Introduction
72 The majority of offshore structures, whether conventional oil and gas platforms or wind turbines, are
73 supported by driven open-ended steel piles, used as single, large diameter laterally loaded monopiles or
74 multiple axially loaded piles for a jacket structure. The piles resist both the topside loads and the
75 environmental wind and wave forces (Arany et al., 2017; Prendergast et al., 2018) and efficient
76 installation is critically important to minimise time delays and prevent material damage. Piles are
77 installed using large hammers, which are usually hydraulically powered to provide a controlled driving
78 force. Prior to selecting an appropriate hammer, a driveability analysis is usually performed to ensure
79 the selected equipment is capable of installing the pile to the target depth in a reasonable time-frame
80 and without overstressing the steel pile shaft. This process is essential to the smooth installation of any
81 offshore structure, as driving delays can result in significant financial overspends due to vessel down-
82 time. Premature refusal or structural damage to the piles can also threaten the feasibility of an offshore
83 project. Therefore, a comprehensive driveability analysis should be undertaken that considers the entire
84 driving system including the hammer performance, pile geometry, site specific soil conditions and the
85 soil-structure interaction problem.

86 The Static Resistance to Driving (SRD) is a profile of shaft and toe resistance developed during pile
87 installation, and an estimate of this is required to perform a driveability study. An SRD profile differs
88 from a static capacity profile in that it models the cumulative increase in shaft capacity with further pile
89 penetration and has a toe resistance associated with each driving increment, as opposed to a static profile
90 with a single base resistance. Moreover they differ in terms of time, degree of mobilisation and
91 consolidation. Accurately predicting the soil-structure interaction is critical to the driveability process
92 and is arguably the most challenging aspect (Prendergast and Gavin, 2016; Wu et al., 2018). Traditional
93 approaches for predicting SRD such as Stevens et al. (1982), Toolan and Fox (1977), and Semple and
94 Gemeinhardt (1981) are largely empirical and therefore, extrapolation to pile geometries and soil
95 conditions outside of the dataset on which they are based is highly questionable. The application of such
 Published in Ocean Engineering 166 2018 pp.76-91

96 methods in the extreme hard tills or very dense sands (where the CPT end resistance qc value is typically
97 in the range 30 – 100 MPa) in the North Sea and for the large diameter monopiles supporting wind
98 turbines should be assessed. Aldridge et al. (2010) highlights the difficulties in predicting the
99 installation resistance of 2.6m diameter piles at the Clair platform in the North Sea, west of the
100 Shetlands, where the undrained shear strength of the underlying glacial till exceeded 2500 kPa. In this
101 instance, a series of pile driving trials were conducted in advance of the platform installation to ensure
102 the piles could reach the target depth. Predictions based on the IC-05 (Jardine et al., 2005) methodology
103 which accounts for high strength of the clay, the pile geometry, group effects and cyclic loading
104 provided much more consistent estimates of the pile resistance than standard offshore methods.

105 Schneider and Harmon (2010) proposed a pile driveability model based on the UWA-05 (Lehane et al.,
106 2005) method that incorporates friction fatigue and accounts for the effects of pile plugging. They note
107 that inertial effects during driving mean that the soil plug remains at or near the sea bed level during
108 installation. As a result they propose a methodology in which the base resistance develops only on the
109 pile annulus, and the shaft resistance develops both on the external pile surface and internally on the
110 pile plug. They suggest that the stress on the pile annulus be taken as 35% of the CPT qc value at the
111 pile tip and the shear resistance mobilised by the soil plug is 50% of the external shear resistance. They
112 found this approach gave consistent predictions of driving resistance for four, open-ended pipe piles
113 driven with diameters ranging from 0.356m to 2m at three sites, in Japan, USA and the Netherlands.

114 Byrne et al. (2012) examined the ability of commonly employed pile driveability models (Semple and
115 Gemeinhardt, 1981; Stevens et al., 1982; Toolan and Fox, 1977) to predict the installation response of
116 a 0.762m diameter skirt pile and a 4.2m diameter monopile installed in dense North Sea sand. Somewhat
117 surprisingly the models provided poorest predictions of the installation response of the 0.762m pile with
118 the range of measured to predicted blow-counts varying from 30% to 180% at the final penetration
119 depth of 34m. The methods provided much closer estimates for the 4.2m diameter pile at the final
120 penetration depth of 31.5m (estimates in the range 55% to 135%). However, for shallow penetrations
121 of the monopile the variance was much larger and methods which under-predicted the blow-counts for
122 shallow penetrations, tended to over-estimate the resistance at depth and vice-versa. This suggests some
123 fundamental inherent bias in the models which would lead to inconsistent estimates of the full pile
124 driving process.

125 In this paper, records from a number of pile installations in the North Sea are used to compare the
126 performance of conventional driveability analyses and new CPT approaches modified to account for
127 important processes including pile ageing, friction fatigue, low base displacement and plugging (Gavin
128 and Lehane, 2007). The pile driving data from six sites, namely Caravel, Cutter, Shamrock, L09FA1,
129 L09FB1 and Skiff are used to compare the predictive performance of the models. The Skiff platform is
130 supported on a jacket with 0.762m diameter piles, whilst the remaining platforms are founded on single,
131 4.2m monopiles. The limitations and biases of each driveability method are assessed and
132 recommendations are made for a more accurate scientific approach.

133

134 2. Modelling Process

135 Determining the optimum pile geometry is an iterative process, where the trade-off between the higher
136 capacities achieved using larger piles is offset by the increased difficulty and risk associated with
137 driving these piles to the desired penetration. A flow-chart describing the principal inputs required and
138 the analysis procedure used to conduct a driveability study is shown in Fig. 1. The three main inputs
139 are the soil parameters, pile properties and hammer details. The pile properties and hammer
 Published in Ocean Engineering 166 2018 pp.76-91

140 characteristics are relatively well understood and can be determined with a reasonable degree of
141 accuracy. Given the difficulties in sampling and testing offshore soils in-situ Cone Penetration Testing
142 (CPT) is widely undertaken for offshore projects with most installations having at least one complete
143 CPT profile. Given the similarities between CPT and pile installation, many correlations have been
144 developed linking pile end, qb and shaft resistance, qs to the CPT end resistance, qc which usually give
145 better predictions of pile response than approaches which use soil parameters within an effective stress
146 framework (Jardine et al., 2005). The main output of interest for this paper is a prediction of the blow-
147 counts required to drive a given pile, as this will be compared to the real blow-counts across the
148 investigated sites. It should be noted that there is potential for several combinations of the input
149 parameters to yield similar output values. The SRD profiles in this paper are derived using measured
150 CPT profiles as input. To ensure that the modelling is as accurate as possible, information from the
151 monitoring reports related to hammer type, input energy, driving delays, etc. for each pile is carefully
152 considered in each driveability analysis to ensure the predictions for each proposed soil resistance are
153 genuine.

154

155 Fig. 1. Flow-chart of principal inputs and outputs available from a wave equation based driveability
156 analysis.

157

158 The total resistance of a pile to driving results from a combination of the static SRD, dynamic increases
159 in pile capacity due to inertial effects and increases in capacity due to viscous rate effects. Driveability
160 methods used to derive the static SRD are discussed in Section 3. The remaining effects (inertia, viscous
161 rate effects) are accounted for within the wave equation analysis problem. In this paper, a commercially
162 available finite-difference software GRLWEAP Off-Shore 2010 (Pile Dynamics, 2010) was used to
163 perform the operations and analyse the energy transferred to the pile from each hammer blow.
164 GRLWEAP is a 1-D wave equation analysis software capable of simulating the response of a pile to
165 pile driving equipment, and is fundamentally based on solving the wave equation shown in Eq. (1)
 δ2u   δ2u 
166   
2 
 E  2  (1)
 δt   δx 
167 where  is the mass density (kg/m3), E is the elastic modulus (N/m2). The wave speed, c is a function
E
168 of the mass density and elasticity as follows, c  .

169
170 GRLWEAP has a number of inbuilt static SRD models, however, it also allows the user manually input
171 shaft and end resistance profiles, thus making it ideal for the problem in this paper. The dynamic forces
172 and viscous rate effects are represented by damping values, which vary according to soil type. It is usual
173 for an SRD model to be used with an accompanying set of standard damping values. Also specified are
174 quake values, the displacement required to achieve yield, see Fig. 2. The parameters used in this study
175 are presented in Table 1 and are derived from original references, where possible.
176

177 Fig. 2. Definition of quake

178
 Published in Ocean Engineering 166 2018 pp.76-91

179 Table 1. Quake and damping values.

180
181 GRLWEAP has an extensive archive of hammer types with a database of their properties (such as
182 hammer masses and drop heights, among other properties). Monitoring reports from each of the
183 installations considered in this paper included details of events such as delays occurring during driving,
184 thereby allowing the hammer performance/efficiency and driving system to be modelled accurately.
185 With the exception of the monopile at Cutter, each pile was dynamically monitored which enabled the
186 pile enthru energy (energy that the pile experiences) to be calculated and replicated in the analysis.
187 Hammer stroke heights were adjusted during the modelling process so that the output energies matched
188 the recorded values. Standard efficiencies were applied based on suggested values for each hammer.
189

190 The friction fatigue effect, whereby shear resistance in a given layer reduces as the pile tip advances is
191 included in the later SRD models, usually by means of including a degradation term in the calculation
192 of the shear stresses of the form (h/R)n where h is the vertical distance from the pile tip to the soil
193 horizon in question and R is the pile radius, see Fig. 3.

194

195 Fig. 3. Definition of (h/R) expression.

196

197 Because the term causes the shape of the shear resistance distribution to change with pile penetration,
198 some averaging technique is necessary. Schneider and Harmon (2010) found that the shape of the shaft
199 friction distribution had a negligible effect on the resultant bearing graph. They suggest the change in
200 shaft capacity between two successive depth increments be used to calculate the pseudo average shaft
201 friction (Δτf,avg), as shown in Eq.(2).
202

203  f ,avg 
QS ,L   QS , L 1
(2)
DL
204

205 where ∑QS,L is the cumulative shaft resistance at a given tip depth; ∑QS,L-1 is the cumulative shaft
206 resistance at the previous depth increment; ΔL is the depth increment; and D is the pile diameter. This
207 pseudo averaging technique was applied to the relevant models incorporating friction fatigue in the
208 present analysis.
209
210 3. Traditional Static Resistance to Driving (SRD) approaches
211 Numerous driveability approaches have been proposed throughout the years to calculate the soil static
212 resistance to driving and are still frequently used in North Sea pile design. The initial methods were
213 developed in the late 1970’s and early 1980’s. Three traditional driveability models were employed in
214 the analysis in this paper.
215
216 3.1 Toolan & Fox (1977)
217 The Toolan and Fox model proposes the calculation of the unit toe resistance, for both cohesive and
218 cohesionless soil, as a weighted average of the cone tip resistance (qc) over a number of pile diameters
219 above and below the pile tip. The unit skin friction in sands can be determined as a fraction of the
 Published in Ocean Engineering 166 2018 pp.76-91

220 recorded cone resistance, (1/300 for a dense sand) or alternatively may be computed in accordance with
221 the American Petroleum Institute (API) (API, 2007) guidelines, and is limited to 120 kPa. For a fully
222 coring pile, unit skin friction is applied to the internal and external shaft area equally while the unit toe
223 resistance is applied to the pile annulus.
224
225 3.2 Stevens et al. (1982)
226 For cohesionless materials, both unit toe and skin resistances are calculated using the standard static
227 capacity procedures outlined in the API method (API, 2007). Limiting input values of unit skin and toe
228 resistance are assigned while the model defines lower (LB) and upper bound (UB) static predictions for
229 plugged and coring conditions. Large open-ended pipe piles, similar to those installed at the six
230 locations considered in this paper, usually remain fully coring during pile installation (internal soil core
231 level approximately at external sea bed level). The lower bound case adopts an internal shaft friction
232 half that of the exterior, with the upper bound assuming that both are equal. The unit skin friction is
233 first calculated from the API method (API, 2007) and adjusted incrementally by a capacity factor
234 determined empirically from wave equation analysis.
235
236 3.3 Alm & Hamre (2001)
237 The original Alm and Hamre (1998) model was developed from back-calculated driveability studies
238 from North Sea installations. An updated version presented in 2001 moved to a CPT based approach to
239 address issues with variability and uncertainty in selection of soil parameters. The model benefited from
240 an enlarged database containing longer and larger (1.8-2.7m) diameter piles and incorporated the
241 friction fatigue effect.

242 The ultimate shaft friction, f is given by Eq.(3).

243  f   res  τ f max   res e  kh (3a)

0.13
 ' 
244 τ f max  0.0132qc  v 0  tan  (3b)
 Patm 
0.5
 q 
245 k  0.0125 c  (3c)
 Patm 

246 Where: fmax is the peak shaft friction, res is the residual friction (= 0.2 fmax), h is the depth of the layer
247 from the pile tip, k is a shape factor and  is the interface friction angle. The unit end bearing resistance,
248 qb is given by Eq. (4).
0.2
 q 
249 qb  0.15qc  c  (4)
  'v 0 

250

251 4. Application of axial static capacity approaches to driveability

252 The CPT based design methods for calculating the axial static resistance of piles in sand known as the
253 IC-05 and UWA-05 methods drew heavily on the findings from highly-instrumented model pile tests
254 undertaken by a number of researchers at Imperial College London (Jardine et al., 2005; Lehane et al.,
255 2005). In particular these tests provided new insights into the mechanisms controlling the development
 Published in Ocean Engineering 166 2018 pp.76-91

256 of shaft friction for displacement piles in sand. The IC-05 method for estimating the ultimate shaft
257 resistance is shown in Eq.(5).
258
 0.13
  'v 0   0.38 
259  f  a 0.029  b  qc   max  h ,8    ' rd  tan  f (5)
   R *  
  pref   
260 Where a = 0.9 for open-ended piles, b = 1.0 for piles in compression, R* is the equivalent radius of a
261 closed-ended pile (assuming no plugging occurs during installation) and σrd is the change in radial
262 stress due to interface dilation. This term can be ignored for offshore piles.
263
264 The unit base resistance is given by Eq.(6).
265
qb  D 
266   0.5  0.25 log  (6)
qc  DCPT 
267
268 Where: DCPT is the diameter of the CPT penetrometer.
269
270 The UWA-05 method has a similar formulation for shaft resistance as shown in Eq. (7).
271
f  0.3   h 
0.5

272 f  t 0.03  qc  Ar .eff max  ,2    ' rd  tan  f (7)
fc    D  
2
𝑓  Di 
273 Where: 𝑓𝑡 = 1 for compression and 0.75 for tension. The effective area ratio, Ar ,eff  1  IFR   ,D
𝑐
D
274 and Di are the external and the internal diameter of the pile respectively. IFR is the incremental filling
275 ratio (or change in soil plug length for an increment of pile penetration).
276
277 The unit base resistance is calculated using the expression in Eq.(8).
278
279 qb0.1 / qc,avg = 0.15 + 0.45 Ar,eff (8)
280
281 While not explicitly designed for determining a soil SRD, similarities between the mechanisms
282 controlling installation resistance and static capacity suggest that with modification and adaptation these
283 methods could prove appropriate for use in driveability analyses. A case study presented by Overy and
284 Sayer (2007) indicated that the IC-05 method gave reasonable predictions for drill-drive operations of
285 the main conductor piles at the same Skiff site considered in this study.
286
287 In this paper the possible contribution of four factors that could differentiate the installation and static
288 loading processes are considered:
289
290 (i) The phenomenon of pile ageing (capacity increase with time after installation) is becoming more
291 widely accepted. The recent CPT based design methods are calibrated using load tests performed
292 generally 10 to 30 days after installation. Therefore, the pile resistance during installation will be
293 lower than the models suggest.
294 (ii) Given the relatively low displacements experienced during individual hammer blows the rate of
295 pile base mobilisation is considered explicitly.
 Published in Ocean Engineering 166 2018 pp.76-91

296
297 The behaviour of the soil plug during installation and under static loading can differ substantially. Due
298 to inertial effects during driving the internal soil plug resistance is low and large diameter piles usually
299 remain near fully coring (the soil plug moves inside the pile at a similar rate to the pile penetration). In
300 SRD and static capacity models two failure conditions are considered, for static loading: (i) a plugged
301 failure occurs when the soil beneath the pile tip displaces and the inner column of soil remains in contact
302 with the inner pile surface, advancing with the downward moving pile under failure, in effect almost
303 akin to a closed ended pile, (ii) an unplugged failure, where the internal soil cylinder remains stationary
304 as the pile progresses. In reality, both mechanisms may take place during failure.
305
306 The IC-05 method has a procedure and guidance to predict whether plugged or unplugged failure
307 develops. In the instance of unplugged failure (which was the mode of failure predicted for all piles in
308 this study) the full CPT resistance is applied over the annular base area only and the plug resistance
309 (internal shaft friction) is not taken into account explicitly. For the UWA-05 model, the plug is
310 presumed not to fail during static loading and a modified unit base resistance (which considers the
311 effective area ratio and accounts directly for the degree of plugging during installation) acts over the
312 gross pile area. This takes into consideration the reduced base stiffness developed during coring
313 installation of the pile. While these may provide accurate static capacity estimates, it may not model or
314 represent the interaction between the pile and soil plug during driving, as noted by Overy and Sayer
315 (2007) for North Sea pile installations.
316
317 The base capacity evaluated from the IC-05 and UWA-05 models assume pile tip displacements (wb)
318 of 10% of the pile diameter (D), as the failure criteria. During driving, the pile penetration per blow is
319 much lower than this value and a reduction factor should be applied to account for this effect. A three-
320 stage base resistance-settlement model, proposed by Gavin and Lehane (2007), is implemented as a
321 means of estimating the base resistance mobilised during each hammer impact. The model,
322 schematically presented in Fig. 4, considers the pile tip displacement (wb), normalised by the pile
323 diameter (D), and plotted against the mobilised base resistance (qb).
324
325 Fig. 4. Base resistance-settlement model (Gavin and Lehane, 2007)

326

327 The base resistance-settlement model considers the initial settlement response to be linear until a yield
328 strain (wby/D) is reached (assumed to occur at 1.5% of pile diameter), followed by a non-linear parabolic
329 stage to a strain at 10% of pile diameter (i.e. wb/D of 0.1). The linear stage (wb/D < wby/D) is governed
330 by the small strain soil elastic stiffness (E0). While E0 should be computed from shear wave velocity
331 measurements based on seismic cone or bender element tests, in practice, these may not be available
332 and it can be approximated using correlations with CPT qc data (Prendergast et al., 2013), such as those
333 reported by Robertson (1990) and Schnaid et al. (2004). The linear portion of the curve can be
334 represented by Eq.(9).
335
  w 
336 qb  k  b   qb ,res (9a)
  D 
 4  E 
337 -and- k    0 2  (9b)
   1  
 Published in Ocean Engineering 166 2018 pp.76-91

338
339 where ν is the Poisson’s ratio. The parabolic portion (wby/D < wb/D < 0.1) is given in Eq.(10):
340
  wby 1n  w  n 
341 qb  k    b    qb,res (10)
  D   D  
342
343 The residual base stresses (qb,res) associated with previous hammer blows may be of the order of 1% to
344 10% of the CPT qc value at a given depth and the sensitivity of the analysis to a range of potential
345 residual stress conditions is assessed in this paper. The base-settlement model was implemented as a
346 modified form of the UWA-05 approach with the wb/D ratio being estimated based on a typical
347 displacements recorded for piles installed at the sites considered.
348
349 5. Pile Database
350 The installation database assembled by Shell UK was interrogated and the results from the installation
351 of five monopiles were selected for analysis. These monopiles are 4.2m diameter steel open ended piles
352 installed at Caravel, Shamrock, L09FA1, L09FB1 and Cutter, respectively. In addition, a 0.762m
353 diameter skirt pile supporting a jacket structure at the Skiff site was also analysed for comparative
354 purposes. The locations of each of the installations are shown in Fig. 5 and Table 2. The piles were
355 predominantly driven in medium dense (≈40% overall) to dense sand (≈20%) with frequent very dense
356 zones (≈30%). There were instances of loose to very loose sand (<10%) and layers and bands of clay
357 were occasionally present.
358

359 Fig. 5. Location map showing pile locations off Dutch and UK coasts.

360

361 Table 2 compares the primary geotechnical properties at each site and provides other relevant
362 information such as results from laboratory tests conducted during the site investigation. The sites are
363 broadly similar with some expected variation in sand relative densities and CPT qc profiles at given
364 depths.
365
366 Soil plug measurements were recorded at sea bed level for Cutter after driving ceased and although no
367 soil plug measurements were made at the remaining sites, the monopiles would most likely have been
368 fully coring. It is also reasonable to assume that the Skiff skirt pile would be coring (or partially
369 plugged), according to IC-05 plugging guidelines.
370

371 Table 2. Site Description.

372

373 Geotechnical logs, in addition to in-situ and laboratory test reports, were available for each location and
374 provided a comprehensive catalogue of the site conditions from which soil profiles could be derived.
375 Relative densities were evaluated by interpreting the CPT cone resistance based on the relationship
376 proposed by Jamiolkowski et al. (1988) shown in Eq. (11).
 Published in Ocean Engineering 166 2018 pp.76-91

 1 qc 
377 Dr  100 ln 0.51 
(11a)
 2.93 205 'm 

378 where Dr is the estimated relative density (%), qc is the measured cone resistance (kPa), σ'm is the
379 estimated mean effective stress at the test depth (kPa) and
380
 p ' 1  2 K 0  
381  'm   0  (11b)
 3 
382
383 where p0’ is the effective overburden pressure (kPa) and K0 is the coefficient of lateral earth pressure.
384 Tentative upper and lower bound values of K0 were used to estimate the relative density profiles which
385 are illustrated in Fig. 6 and are accompanied by the corresponding CPT qc profiles from each site. The
386 consistency of the sand can be categorised according to the ranges of relative densities proposed by
387 Lambe and Whitman (1969) as presented in Table 3.
388
389 Table 3. Relative Density (Dr) definitions (Lambe and Whitman, 1969)

390

391 When the Dr traces abruptly return to zero percent, this indicates the transition from a sand layer to a
392 clay layer and likewise, where the relative density spikes sharply from zero, this indicates the return to
393 a sand layer. The resistance of stiff-hard clays to an advancing CPT cone is usually less than that of
394 medium dense to dense sand and it would be expected that the measured total cone resistance in clay
395 would be noticeably less than that of the sand. In this way, the two traces should mimic each other at
396 the interface zones between clay and sand. This is mostly true for the sites in Fig. 6. Each 0.5m depth
397 increment used in the calculation of the respective methods’ SRDs had an associated CPT qc value. A
398 simplified profile (ignoring the extremes in CPT qc) was not assumed as this would be un-conservative
399 because of the potential to underestimate the soil strength. Instead, values close to the actual CPT qc
400 traces were assigned in each case.

401

402 Fig. 6. Site CPT qc and Relative Density (Dr) Profiles.

403
404 6. Results
405 In this section, the results are presented as the application of the various methods to the driveability
406 problem, namely traditional approaches (section 6.1), static capacity approaches (section 6.2) and
407 modified static capacity approaches (section 6.3). The performance of the various approaches to
408 analysing monopile driveability is critically evaluated and discussed.
409
410 6.1 Traditional driveability approaches
411 The pile installation performance was predicted by performing wave equation analyses using the
412 driveability software. Having entered the pile properties and SRD profiles the driving process was
413 simulated by adjusting hammer stroke heights and a profile of blow-counts with depth was predicted.
414 Blow count profiles with depth from two of the six sites considered namely; Skiff (pile diameter
415 0.762m) and Caravel (pile diameter 4.2m) are shown in Fig. 7. These types of plot are frequently
416 produced during the pile design phase to determine if excessive blows would be required to drive a pile.
417 It is common in practice to have best and upper estimates. Usually, a nominal number of 250 blows per
418 0.25m penetration is defined as pile refusal, however, this limit can be substantially lower for large
 Published in Ocean Engineering 166 2018 pp.76-91

419 diameter monopiles, due to the increased risk of damage associated with the use of larger hammers
420 required to drive the piles. It is evident in Fig. 7(a) that a number of the approaches predicted more than
421 250 blows/0.25m were required to drive the piles at Skiff.

422 For example the upper bound estimates for the established approaches of Stevens et al. which assumes
423 the pile is plugged predicts refusal at 27.5m. In reality the pile was installed in a partially plugged mode
424 and Stevens et al. (1982) recommend using the Plugged Lower Bound (Steve Plug LB) prediction as a
425 best estimate for piles in dense sands, even for piles which have a high probability of coring during
426 installation. The lower bound line here is in reasonably good agreement with the measured blow counts,
427 albeit underestimating the pile resistance for pile penetrations in excess of 25m. In contrast, when this
428 method was applied to the large diameter monopile, refusal was predicted within the first few metres of
429 penetration. A similar trend was noted for all the monopiles considered. The Stevens approach was
430 developed using a dataset of installations consisting mainly of piles of less than 2m in diameter,
431 typically used during the 1970’s and 1980’s. While the approach provides an acceptable estimate here
432 for the 0.762m diameter skirt pile at Skiff it is totally inadequate for the monopiles, where refusal was
433 almost immediate. This highlights the potential pitfalls in extrapolating existing methods to piles of
434 larger diameters.

435 Although the Toolan & Fox (T&F) and Alm & Hamre Upper Bound (A&H UB) both predicted early
436 refusal for the pile at Skiff, they return to within acceptable limits at deeper penetrations. It is at the
437 engineer’s discretion and best judgement must be used in deciding whether these periods of perceived
438 hard driving would be acceptable during the course of a driveability study, considering that it is the
439 upper bound estimate of Alm & Hamre that predicts this. The best estimate (A&H Best) falls
440 comfortably below 250 blows/0.25m penetration.

441 For the methods that assume the pile is coring through installation, e.g. Stevens et al. (Stevens Cored
442 LB & UB), the predicted blow counts increased steadily with penetration for both the Skiff and Caravel
443 piles and were not overly sensitive to abrupt changes in soil profile. The estimates tend to under predict
444 the recorded blow counts at each site. It is somewhat unexpected to see the high degree of variation in
445 the predictions for Skiff in comparison to the relatively narrow band produced at Caravel, excluding
446 the Stevens plugged estimates. It is worth noting that the soil profile at Skiff has a much higher
447 proportion of dense to very dense sand over the embedded pile length that may influence the disparity
448 in the predictions. Table 4 shows the total blow-counts required to drive the piles as estimated from
449 each method, for both Skiff and Caravel.

450

451 Table 4.Total blow-counts measured and predicted for Skiff and Caravel

452

453 Fig. 7. Blow count predictions vs measured, (a) Skiff, (b) Caravel

454

455 In addition to comparing blow counts it is important to consider both the driving stresses and installation
456 time to ensure that the pile is not damaged and the installation costs are not excessive. This is
457 particularly important for wind farm sites where a large number of piles need to be driven. While
458 neither of these aspects were explicitly considered in the present study, it is noteworthy that moderately
 Published in Ocean Engineering 166 2018 pp.76-91

459 high but acceptable driving stresses can become an issue with fatigue damage, if prolonged periods of
460 high blow counts are expected, so an indirect measure of this is provided.

461

462 6.2 Unmodified static capacity approaches

463 In this section, the application of more recently developed CPT-based static (axial) capacity approaches
464 to estimating driveability is investigated. The IC-05 and UWA-05 static capacity methods are first
465 applied in their raw (unmodified) form to estimate driveability, by deriving SRD profiles for the shaft
466 and toe resistance based on each method (Jardine et al., 2005; Lehane et al., 2005). The friction fatigue
467 effect, which results in the distribution of shaft friction varying with advancing pile tip, is incorporated
468 by calculating the pseudo-average shaft friction between adjacent depths, see Eq. (2). The results of the
469 driveability analysis for the six sites derived using the IC-05 and UWA-05 and compared to the recorded
470 blow counts are shown in Fig. 8.
471
472 Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches

473
474 From the results in Fig. 8, it can be observed that for Skiff, Caravel, Cutter and Shamrock, both methods
475 broadly over-predict the blow counts required to install each pile. For Caravel and Shamrock, the IC-
476 05 method provides a reasonable estimate of the driveability with some significant over-prediction in
477 places (e.g. Caravel upper layers). However, for L09FA1 and L09FB1, the IC-05 approach tends to
478 under-predict somewhat. The relatively poor predictive resistance of these methods in their unmodified
479 form is unsurprising as they were developed to predict the medium-term (10 to 30 day following
480 installation) static capacity of piles. Moreover, the base resistance calculated from the IC-05 and UWA-
481 05 approaches assumes a base displacement of 0.1D is mobilised. Nonetheless, the methods appear to
482 provide reasonable profiles on the driveability resistance (mirroring the blow counts in the different
483 layers) and therefore the next section considers some modifications that could be applied to the methods
484 to simulate the actual driving process more closely.
485
486 6.3 Modified static capacity approaches
487 The IC-05 and UWA-05 static capacity methods are modified in this section and applied to estimating
488 driveability. In the first instance, to account for the fact that pile ageing effects are inherently considered
489 in the static capacity approaches, the shaft capacity as derived by each method is reduced to 70% of the
490 unmodified value. The reduction factor of 0.7 for short-term shaft capacity was chosen based on the
491 interpretation by Jardine et al. (2006) of Intact Ageing Curves from pile tests that investigated the ageing
492 characteristics of piles installed in dense sand at Dunkirk. Based on static and dynamic load tests, the
493 authors proposed that the end of driving shaft capacity of piles was 70% of the resistance predicted
494 using the ICP design method. This proposal was consistent with field tests reported by Gavin et al.
495 (2013) and Karlsrud et al. (2014) summarised in Gavin et al. (2015). For the UWA-05 approach, the
496 base resistance is further modified herein using the base-settlement model in Fig. 4. In this modification,
497 a revised base resistance is calculated using the actual settlement under each hammer blow (averaged
498 into layers) and normalized to the pile diameter in each case. The new base resistance is calculated
499 using Eq. (9), initially assuming no residual base resistance (qb,res) acts on the system. The results of the
500 modified approaches with the actual measured blow counts are shown in Fig. 9.
501
502 Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches

503
 Published in Ocean Engineering 166 2018 pp.76-91

504 In Fig. 9, a more reasonable prediction of blow counts is achieved in each case. The UWA approach
505 predicts a significantly lower blow count vs penetration than the unmodified approach in Fig. 8. Broadly
506 speaking, it now under predicts the blow counts for each pile. The IC-05 approach both over and under
507 predicts the blow-counts across the sites, though generally speaking provides a reasonable estimate in
508 most cases. This approach under-predicts the response at the L09FA1 and L09FB1 sites. The blow-
509 counts in the upper layers of the Cutter site are heavily over predicted. This is a result of the estimated
510 w/D under each hammer blow in the upper layers being over-predicted due to the lack of actual blow
511 count information for these depths (an average was assumed based on the initial blow-count
512 information). Therefore, the results for Cutter in the upper layers are not a realistic estimate of the actual
513 behaviour and can be ignored. Results from the Cutter site are less reliable overall due to a failure of
514 the logging system reading the blows for depths larger than 17m bgl, as can be seen by the stepped
515 nature of the data signal.
516
517 The modified UWA approach provides a reasonably consistent under prediction of blow counts across
518 all sites. As the base resistance was developed using the resistance-settlement model in Fig. 4, and the
519 presence of potential residual base stress after each hammer blow is initially ignored, it is reasonable to
520 account for the presence of potential residual base stress adding to the overall resistance properties. The
521 impact of residual loads on piles in sand is well known however, measurements of residual loads on
522 open-ended piles are rare. Paik et al. (2003) report residual loads measured on 356 mm diameter closed
523 and open-ended piles installed in dense sand. They found that the residual stress (qb,res) normalised by
524 qc was similar for both piles (in the range 11% to 14% of qc) despite the open-ended pile being nearly
525 fully coring, with a final incremental filling ratio (IFR) of ≈ 78%. Kirwan (2015) reports residual load
526 measurements on a 340 mm diameter open-ended pile installed in dense sand. The residual load
527 appeared to be affected by the IFR with a ratio qb,res/qc of 27% at the end of installation when the IFR
528 was 40%. Residual base stresses varying from 1% of the CPT qc tip resistance to 10% qc are assumed
529 in a parameter study, conducted herein.
530
531 The modified toe resistance from the UWA approach was calculated using the linear portion of the
532 base-resistance settlement model from the actual settlement per blow information from each site
533 averaged into layers. In order to investigate the effect of residual base stresses potentially existing after
534 each hammer blow, residual stresses of qc where   0.01 0.02 0.05 0.08 0.1T were added
535 to the modified toe resistance profile for each site. The resistance was added in a step-wise approach
536 and only added once there was sufficient negative skin friction present at a given depth to resist the
537 residual stress (i.e. the pile was in equilibrium after each hammer blow). The negative skin friction in
538 tension was assumed to be 80% of the compression resistance. The derived modified toe resistance SRD
539 profiles with step-wise residual stresses and the predicted blow counts for the Caravel site are shown in
540 Fig. 10.
541
542 Fig. 10. Effect of residual base stress at Caravel site, (a) UWA modified to resistance with varying
543 residual stresses added, (b) Predicted blow counts for a range of residual added stresses

544

545 Fig. 10(a) shows the derived modified UWA-05 toe resistance SRD profiles for Caravel with step-wise
546 residual base stress added in increments corresponding to 0.01qc, 0.02 qc, 0.05qc, 0.08qc and 0.1qc at
547 each depth. Fig. 10(b) shows the blow count predictions for the various input profiles. The recorded
548 blow counts are also shown on the plot. A reasonable match to the recorded profile is obtained with
549 the step-wise residual stress added up to 10% of the qc profile. For the remaining sites, the results of a
 Published in Ocean Engineering 166 2018 pp.76-91

550 similar analysis are summarised herein. Adding the residual stress component leads to a reasonable
551 prediction for Cutter (though recorded values are albeit somewhat unreliable due to logger failure) as
552 well as for the 0.762m pile at Skiff. The UWA modified approach with added residual stress still leads
553 to an under prediction for L09FA1, L09FB1 and Shamrock sites.
554
555 The modified IC-05 method previously only considered a reduction in the shaft friction during driving
556 to remove the ageing effect with the toe resistance remaining unmodified. This approach provided a
557 reasonable estimate of the blow counts for the Shamrock site, see Fig. 9(c). A brief analysis is conducted
558 herein to investigate the effect of adopting the UWA-05 modified base resistance with the IC-05
559 modified shaft resistance and applying the residual toe resistances to the case of the Shamrock site. The
560 results are shown in Fig. 11, which show an under prediction in each case though the general shape of
561 the blow count with depth is reasonably represented, especially by the step-wise addition of the base
562 stress up to 10% of the qc tip resistance. This analysis suggests that the addition of a residual base
563 resistance component to the modified toe resistance profile provides an improved prediction of the blow
564 counts. Overall, however, the modified IC-05 method with reduced shaft friction and unmodified base
565 resistance seems to provide the most reliable method for estimating blow counts in the context that it
566 mainly provides a conservative estimate in a driveability context, except for L09FA1 and L09FB1
567 where it under-predicts. The modified UWA-05 approach primarily under-predicts the blow counts for
568 the given sites tested, though the addition of residual base stress reduces this under-prediction
569 considerably in many cases.
570
571 Fig. 11. Predicted and recorded blow counts for Shamrock with residual added stress and modified
572 base IC-05 method.

573

574 7. Conclusions

575 This paper examined the feasibility of applying both existing pile driveability approaches and
576 modifying static capacity approaches to monopile foundations installed in North Sea sand. No single
577 method investigated proved reliable for predicting blow counts across all the sites considered. For the
578 traditional approaches, there is a tendency to underestimate the blows for sites with medium dense to
579 dense sand. An exception being the use of the Stevens et al. plugged approach, which should not be
580 applied for large diameter piles. The recommendation of using the lower bound prediction of the
581 plugged case in dense sands applied for smaller diameter piles is clearly not suitable for monopiles.

582 Applying axial static capacity approaches (IC-05 & UWA-05) in their raw unmodified form is mostly
583 unsuitable for monopile driveability, with substantial overestimates frequently produced. This is
584 particularly true for the UWA-05 approach, which vastly over-predicted blow counts in all cases. This
585 is a result of the method taking an average base resistance (accounting for plugged behaviour during
586 static loading) acting across the entire pile base area. The IC-05 approach over-predicts the blows for
587 Skiff, Cutter and the upper layers of Caravel while providing a somewhat reasonable estimate for
588 Caravel lower layers and Shamrock. The IC-05 approach generally under-predicted the blow counts at
589 L09FA1 and L09FB1. Adopting the base settlement model in the UWA-05 approach provides a
590 significant improvement, albeit the predictions tend to fall short of the recorded blow counts. The
591 inclusion of potential residual base stresses of up to 10% of the qc value at a given depth gave a further
592 improvement to the model that is keeping with field measurements, though still resulted in an under-
593 prediction in each case. The modifications applied, while not a formal method for predicting
594 driveability, should certainly be considered as a guide in estimating the blow counts for monopiles.
 Published in Ocean Engineering 166 2018 pp.76-91

595 Comparisons between the monopiles and skirt pile are difficult to evaluate as only a single, isolated
596 case was considered at Skiff. Overall, however, the IC-05 with modified shaft friction and the UWA-
597 05 with modifications including additional residual base stresses provided reasonable estimates of the
598 blow counts for the cases considered, and would certainly provide an estimate of the expected behaviour
599 of a monopile under driving to an acceptable industry tolerance. If adopted in tandem, they may provide
600 an envelope of the expected response, with perhaps a tendency toward under-prediction as per L09FA1
601 and L09FB1. As with all approaches, engineering judgement should be taken when choosing a method
602 and the analysis in this paper aims to highlight the potential disparity in predictions from the various
603 available approaches.

604 The study undertaken in this paper has looked at the applicability of the various approaches to
605 monopiles with diameters of 4.2m. A future study is recommended on larger diameter monopiles to
606 further investigate the remaining uncertainties associated with these emerging geometries.

607

608 Acknowledgements
609 The authors wish to express their gratitude to Robert Overy and Shell UK for supplying records of the
610 installations. Mainstream Renewable Power are also thanked for their support.
611
612 References

613 Aldridge, T.R., Carrington, T.M., Jardine, R.J., 2010. BP Clair phase 1–Pile driveability and capacity
614 in extremely hard till. Front. Offshore Geotech. II 477.

615 Alm, T., Hamre, L., 2001. Soil model for pile driveability based on CPT interpretations, in:
616 Proceedings of the 15th International Conference On Soil Mechanics and Foundation
617 Engineering. Istanbul, Turkey, pp. 1297–1302.

618 API, 2007. API RP2A-WSD, Recommended Practice for Planning, Designing, and Constructing
619 Fixed Offshore Platforms–Working Stress Design, American Petroleum Institute, Washington,
620 DC.

621 Arany, L., Bhattacharya, S., Macdonald, J., Hogan, S.J., 2017. Design of monopiles for offshore wind
622 turbines in 10 steps. Soil Dyn. Earthq. Eng. 92, 126–152. doi:10.1016/j.soildyn.2016.09.024

623 Byrne, T., Doherty, P., Gavin, K., Overy, R., 2012. Comparison of Pile Driveability Methods In North
624 Sea Sand, in: Proc. 7th Intl Offshore Site Investigation and Geotechnics Conference: Integrated
625 Geotechnologies – Present and Future. Society for Underwater Technology (SUT), London, UK,
626 pp. 481–488.

627 Gavin, K., Igoe, D., Kirwan, L., 2013. The effect of ageing on the axial capacity of piles in sand. Proc.
628 Inst. Civ. Eng. - Geotech. Eng. 166, 122–130.

629 Gavin, K., Jardine, R.J., Karlsrud, K., Lehane, B.M., 2015. The Effects of Pile Ageing on the Shaft
630 Capacity of Offshore Piles in Sand, in: Proceedings of the International Symposium Frontiers in
631 Offshore Geotechnics (ISFOG). Oslo.

632 Gavin, K.G., Lehane, B.M., 2007. Base Load-Displacement Response of Piles in Sand. Can. Geotech.
633 J. 44, 1053–1063.
 Published in Ocean Engineering 166 2018 pp.76-91

634 Hirsch, T.J., Carr, L., Lowery, L.L., 1976. Pile Driving Analyses-Wave Equation User Manual. TTI
635 Prog. Implement. Packag. I-IV.

636 Jamiolkowski, M., Ghionna, V.N., Lancellotta, R., Pasqualini, E., 1988. New Correlations of
637 Penetration Tests for Design Practice, in: Penetration Testing 1988: Proceedings of the First
638 International Symposium on Penetration Testing ISOPT-1. Orlando, FL.

639 Jardine, R., Standing, J., Chow, F., 2006. Some observations of the effects of time on the capacity of
640 piles driven in sand. Geotechnique 55, 227–244.

641 Jardine, R.J., Chow, F.C., Overy, R.F., Standing, J., 2005. ICP Design Methods for Driven Piles in
642 Sands and Clays. London.

643 Karlsrud, K., Jensen, T.G., Wensaas Lied, E.K., Nowacki, F., Simonsen, A.S., 2014. Significant
644 ageing effects for axially loaded piles in sand and clay verified by new field load tests, in:
645 Proceedings of the Offshore Technology Conference. Houston, Texas.

646 Kirwan, L., 2015. Investigation into Ageing Mechanisms for Axially Loaded Piles in Sand. University
647 College Dublin.

648 Lambe, T.W., Whitman, R.V., 1969. Soil Mechanics. John Wiley & Sons.

649 Lehane, B.M., Schneider, J.A., Xu, X., 2005. The UWA-05 method for prediction of axial capacity of
650 driven piles in sand. Front. Offshore Geotech. ISFOG 683–689.

651 Overy, R., Sayer, P., 2007. The use of ICP design methods as a predictor of conductor drill-drive
652 installation, in: Proceedings of the 6th International Offshore Site Investigation and Geotechnics
653 Conference: Confronting New Challenges and Sharing Knowledge . Society for Underwater
654 Technology (SUT), London, pp. 333–340.

655 Paik, K., Salgado, R., Lee, J., Kim, B., 2003. Behavior of Open- and Closed-Ended Piles Driven Into
656 Sands. J. Geotech. Geoenvironmental Eng. 129.

657 Pile Dynamics, 2010. GRLWEAP Software.

658 Prendergast, L.J., Gavin, K., 2016. A comparison of initial stiffness formulations for small-strain soil
659 – pile dynamic Winkler modelling. Soil Dyn. Earthq. Eng. 81, 27–41.
660 doi:10.1016/j.soildyn.2015.11.006

661 Prendergast, L.J., Hester, D., Gavin, K., O’Sullivan, J.J., 2013. An investigation of the changes in the
662 natural frequency of a pile affected by scour. J. Sound Vib. 332, 6685–6702.
663 doi:http://dx.doi.org/10.1016/j.jsv.2013.08.020i

664 Prendergast, L.J., Reale, C., Gavin, K., 2018. Probabilistic examination of the change in
665 eigenfrequencies of an offshore wind turbine under progressive scour incorporating soil spatial
666 variability. Mar. Struct. 57, 87–104. doi:10.1016/j.marstruc.2017.09.009

667 Robertson, P., 1990. Soil classification using the cone penetration test. Can. Geotech. J. 27, 151–158.

668 Schnaid, F., Lehane, B.M., Fahey, M., 2004. In situ test characterisation of unusual geomaterials, in:
669 Proceedings of the International Conference of Site Characterisation. Porto, Portugal, pp. 49–73.
 Published in Ocean Engineering 166 2018 pp.76-91

670 Schneider, J., Harmon, I., 2010. Analyzing Drivability of Open Ended Piles in Very Dense Sands. J.
671 Deep Found. Inst. 4, 32–44.

672 Semple, R., Gemeinhardt, J., 1981. Stress history approach to analysis of soil resistance to pile
673 driving, in: Offshore Technology Conference. Houston, USA.

674 Stevens, R., Wiltsie, E., Turton, T., 1982. Evaluating Drivability for Hard Clay, Very Dense Sand,
675 and Rock, in: Offshore Technology Conference. Houston, USA.

676 Toolan, F.E., Fox, D.A., 1977. Geotechnical planning of piled foundations for offshore platforms.
677 Proc. Inst. Civ. Eng. 62.

678 Wu, W.H., Prendergast, L.J., Gavin, K., 2018. An iterative method to infer distributed mass and
679 stiffness profiles for use in reference dynamic beam-Winkler models of foundation piles from
680 frequency response functions. J. Sound Vib. 431C, 1–19.

681
 Published in Ocean Engineering 166 2018 pp.76-91

682
683 Fig. 1. Flow-chart of principal inputs and outputs available from a wave equation based driveability
684 analysis.

685
 Published in Ocean Engineering 166 2018 pp.76-91

686
687 Fig. 2. Definition of quake

688
 Published in Ocean Engineering 166 2018 pp.76-91

689
690 Fig. 3. Definition of (h/R) expression.

691
 Published in Ocean Engineering 166 2018 pp.76-91

Fig. 4. Base resistance-settlement model

 Published in Ocean Engineering 166 2018 pp.76-91

Fig. 5. Location map showing pile locations off Dutch and UK coasts.
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0
CPT
Dr

10
Penetration (m)

20

30

40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (a) Caravel
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0
CPT
Dr

10
Penetration (m)

20

30

40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (b) Shamrock
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0

10
Penetration (m)

20 CPT
Dr

30

40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (c) L09FB1
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0

10
Penetration (m)

20 CPT
Dr

30

40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (d) L09FA1
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0

10
Penetration (m)

20

30

CPT
Dr
40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (e) Cutter
 Published in Ocean Engineering 166 2018 pp.76-91

CPT qc (MPa)
0 40 80
0

10
Penetration (m)

20

30

CPT
Dr
40
0.0 0.5 1.0
Relative Density (Dr)

Fig. 6. Site CPT qc and Relative Density (Dr) Profiles. (f) Skiff
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200 250
0
Stevens Plugged LB&UB
Stevens Cored LB&UB
Alm and Hamre Best&UB
5
Toolan and Fox
Recorded

10
Penetration (m)

15

20

25

30

35

Fig. 7. Blow count predictions vs measured, (a) Skiff

 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200 250
0

10
Penetration (m)

15
Stevens Plugged LB&UB
Stevens Cored LB&UB

20 Alm and Hamre Best&UB

Toolan and Fox
Recorded
25

30

35

Fig. 7. Blow count predictions vs measured, (b) Caravel

 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05
UWA-05
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (a)
Skiff
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05
UWA-05
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (b)
Caravel
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05
UWA-05
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (c)
Shamrock
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0

10
Penetration (m)

15

20

25

30
IC-05
UWA-05
Recorded
35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (d)
Cutter
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 100 200
0

10
Penetration (m)

15 IC-05
UWA-05
Recorded
20

25

30

35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (e)
L09FA1
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0

10
Penetration (m)

15
IC-05
UWA-05
20 Recorded

25

30

35

Fig. 8. Blow counts/ 0.25m for all sites with unmodified CPT-based static capacity approaches (f)
L09FB1
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05 MOD
UWA-05 MOD
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (a) Skiff
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05 MOD
UWA-05 MOD
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (b)
Caravel
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 100 200
0
IC-05 MOD
UWA-05 MOD
Recorded
5

10
Penetration (m)

15

20

25

30

35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (c)
Shamrock
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 100 200
0

10
Penetration (m)

15

20

25

30
IC-05 MOD
UWA-05 MOD
Recorded
35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (d) Cutter
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0
IC-05 MOD
UWA-05 MOD
5 Recorded

10
Penetration (m)

15

20

25

30

35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (e)
L09FA1
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150 200
0

IC-05 MOD
10 UWA-05 MOD
Recorded
Penetration (m)

15

20

25

30

35

Fig. 9. Blow counts/ 0.25m for all sites with modified CPT-based static capacity approaches (f)
L09FB1
 Published in Ocean Engineering 166 2018 pp.76-91

Toe Resistance SRD (kPa)

0 500 1000 1500
0
Toe Resistance
Toe with 0.01qc
Toe with 0.02qc
5 Toe with 0.05qc
Toe with 0.08qc
Toe with 0.1qc
10
Penetration (m)

15

20

25

30

35

Fig. 10. Effect of residual base stress at Caravel site, (a) UWA modified toe resistance with
varying residual stresses added
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 20 40 60 80
0
RECORDED UNMODIFIED
+0.01qc +0.02qc
+0.05qc +0.08qc
5 +0.1qc

10
Penetration (m)

15

20

25

30

35

Fig. 10. Effect of residual base stress at Caravel site, (b) Predicted blow counts for a range of residual
added stresses
 Published in Ocean Engineering 166 2018 pp.76-91

Blows/0.25m
0 50 100 150
0
RECORDED UNMODIFIED
+0.01qc +0.02qc
+0.05qc +0.08qc
5
+0.1qc

10
Penetration (m)

15

20

25

30

35

Fig. 11. Predicted and recorded blow counts for Shamrock with residual added stress and modified
base IC-05 method.
 Published in Ocean Engineering 166 2018 pp.76-91

Table 1. Quake and damping values.

Sand Clay
Quake Damping Quake Damping
Method (mm) (s/m) (mm) (s/m) Reference
Side Toe Side Toe Side Toe Side Toe
Alm and Hamre 2.5 2.5 0.25 0.5 2.5 2.5 0.25 0.5 (Alm and Hamre, 2001)
(2001)
Toolan and Fox 2.5 2.5 0.17 0.5 2.5 2.5 0.66 0.03 (Hirsch et al., 1976)
(1977)
Stevens et al. (1982) 2.5 2.5 0.27 0.5 2.5 2.5 0.1 0.5 (Stevens et al., 1982)
UWA (2005) 2.5 2.5 0.25 0.5 2.5 2.5 0.65 0.5 (Schneider and Harmon,
2010)
IC (2005) 2.5 2.5 0.16 0.5 2.5 2.5 0.65 0.5 GRLWEAP standard
values
 Published in Ocean Engineering 166 2018 pp.76-91

Table 2. Site Description.

Caravel Shamrock L09FB1 L09FA1 Cutter Skiff

North Sea
UK 49/20 UK 49/20 Dutch L09 Dutch L09 UK 49/09 UK 48/20
Location
Pile
Diameter 4.2 4.2 4.2 4.2 4.2 0.762
(m)
Wall
Thickness 50/55/60 50/60/70 50/60 50/60/65/70 50/60 38.1
(m)
Pile
40.5 43 40 43 41 41
Length(m)
Penetration
31.5 34 31 34 29 34
(m)
Water
31 30 22 23.5 32 26
Depth (m)
Hammer MHU-600 MHU-600 MHU-600 MHU-600 IHC S-600 IHC S-90
LOOSE
LOOSE
SAND MED
SAND
initially, DENSE DENSE DENSE - DENSE -
initially,
MED VERY VERY VERY VERY
MED
DENSE - DENSE DENSE DENSE DENSE
DENSE -
Dominant VERY silica fine silica fine to fine to fine to
VERY
Soil DENSE to medium medium medium medium
DENSE
Conditions fine to SAND with SAND with SAND with SAND with
slightly
medium Stiff - Hard Stiff - Hard shell Stiff clay
silty SAND
SAND clay layers clay layers fragments, layers
below 8.5m
below 8m present present clay layers present
with clay
with clay present
layers
layers
Sand
Friction
Angles
28°-30° 27°-30° 29°-31° 26°-30° 28°-30° 28°-31°
(Direct
Shear:
Soil-Steel)
Clay Layer
150-400 30-100 175-300 100-300 75-175 50-400
Su (kPa)
 Published in Ocean Engineering 166 2018 pp.76-91

Table 3. Relative Density (Dr) definitions (Lambe and Whitman, 1969)

Consistency Relative Density Dr (%)

Very Loose 0 – 15
Loose 15 – 35
Medium Dense 35 – 65
Dense 65 – 85
Very Dense 85 - 100
 Published in Ocean Engineering 166 2018 pp.76-91

Table 4.Total blow-counts measured and predicted for Skiff and Caravel

Recorde Steven Steven Steven Steven A& A&H T&F

d s s s s H Upper
Plugge Plugge Cored Cored Best
d LB d UB LB UB
Skiff Total 6729 3609 5667 2188 3271 4405 5147 9157
Blowcoun
ts
Notes - - Refuse - - - Refuse Refuse
s at s at s at
29m 33m 33.5m
Carav Total 3635 - - 1383 1761 2180 2904 2533
el Blowcoun
ts
Notes - Refuse Refuse - - - - -
s s

View publication stats