1 Arctic rock coast responses under a changing climate 1 Michael Lim a *, Mateusz Strzelecki b , Marek Kasprzak b , Zuza Swirad c , Clare Webster d , John 2 Woodward a and Herdis Gjelten e 3 a Engineering and Environment, Ellison Building, Northumbria University, Newcastle Upon 4 Tyne, NE1 8ST, United Kingdom 5 b Institute of Geography and Regional Development, University of Wrocław, pl. Uniwersytecki 6 1, 50-137 Wroclaw, Poland 7 c Department of Geography, Durham University, South Road DH1 3LE, Durham UK. 8 d WSL Institute for Snow and Avalanche Research SLF, Davos, Switzerland 9 e Observation and Climate Department, Norwegian Meteorological Institute, PO Box 43 10 Blindern, NO-0313 Oslo, Norway 11 12 *Corresponding author. Tel.: +44191 243 7094; E-mail address: 13 [email protected](M. Lim). 14 15 Abstract 16 It has been widely reported that Arctic sea ice has decreased in both extent and thickness, 17 coupled with steadily rising mean annual temperatures. These trends have been particularly 18 severe along the rock coast of southern Svalbard. Concerns have been raised over the potential 19 for higher energy storms and longer ice-free open water seasons to increase the exposure of 20 Arctic coasts, and consequently the concentration of infrastructure critical to Arctic community 21 survival, to enhanced rates of erosion. Here we present and apply innovative remote sensing, 22 monitoring and process analyses to assess the impact of recent coastal climatic changes. High 23
32
Embed
Arctic rock coast responses under a changing climate · 84 account for geomorphic behaviour and over scales great enough to assess process responses to 85 climatic changes (Lantuit
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Arctic rock coast responses under a changing climate 1
Michael Lim a*, Mateusz Strzelecki b, Marek Kasprzak b, Zuza Swirad c, Clare Webster d, John 2
Woodward a and Herdis Gjelten e 3
a Engineering and Environment, Ellison Building, Northumbria University, Newcastle Upon 4
Tyne, NE1 8ST, United Kingdom 5
b Institute of Geography and Regional Development, University of Wrocław, pl. Uniwersytecki 6
1, 50-137 Wroclaw, Poland 7
c Department of Geography, Durham University, South Road DH1 3LE, Durham UK. 8
d WSL Institute for Snow and Avalanche Research SLF, Davos, Switzerland 9
e Observation and Climate Department, Norwegian Meteorological Institute, PO Box 43 10
have been used to help scale and orientate the surface model that is subsequently converted 238
back from a coloured point cloud to the original temperature scale as a three-dimensional 239
thermal point cloud (Webster et al. 2018). 240
241
The main aim of the thermal mapping has been to explore relative (rather than absolute) 242
temperatures but point based temperature comparisons between thermal images and contactless 243
infrared laser surface thermometer spot measurements were in close (generally <1°C) 244
agreement for randomly selected points across all layers and a range of heights up the cliff. This 245
variance is in line with the manufacturer stated accuracy for the camera (±2%). 246
The relative variations in temperature were compared with a vertical transect of Schmidt 247
hammer (Proceq N-type Silver Schmidt) readings. The Schmidt hammer records the rebound 248
force of a spring-loaded hammer in order to derive a non-destructive relative indication of rock 249
strength. Readings are taken over a clean (outer algal and weathering crust removed with a 250
grinding stone where necessary) rock surface in a grid pattern, producing an average hardness 251
value from 25 measurements. After Strzelecki et al. (2017), hardness profiles at each site have 252
been collected in vertical zones through the rock coast from the intertidal zone, through the 253
high-water level, to the cliff toe and the cliff face to the cliff top. 254
4. Results 255
4.1 Arctic cliff recession 256
12
The cliff responses recorded lower rates of retreat (Fig. 2) than those originally quantified from 257
the same cliff by Jahn (1961), although it is not currently possible to determine the validity of 258
a quantitative comparison with physically derived historic rates. The highest annual modern 259
cliff recession rate recorded was 0.019 m a-1, which occurred between 2014 and 2015, a period 260
when storm incidence was high and mean annual air temperature was cooler; -2.27 °C relative 261
to -0.82 °C in 2015 - 16 and -1.55 °C in 2016 - 2017 (weather data provided by the Norwegian 262
Meteorological Institute, after Gjelten et al., 2016; data calculated from summer to summer to 263
reflect the survey schedule). The following year of monitoring (2015-2016) recorded an annual 264
recession rate of 0.017 m a-1 and the final year of monitoring (2016-2017) produced a rate 265
almost 50 % lower than the first year (0.010 m a-1). These modern rates are generally in line 266
with the rates seen in coastal rock cliff environments in more temperate coastal settings (see for 267
example Kirk 1977; Rosser et al. 2007; Sunamura 1992). The spatial distribution of change 268
during the highest monitored rates (2014-2015) was concentrated on areas protruding from the 269
cliff face such as cliff tops, spurs and overhangs. These areas are potentially more exposed to 270
high energy events, stress concentrations and mechanical action associated with snow, ice-foot 271
and ice floe processes. 272
273
The distribution of failures has been further analysed through the relationship between the 274
numbers of different sized failures that occurred from the rock cliffs (Fig. 2). The volume-275
frequency relationships for all years of monitoring follow power law distributions. The 276
exponents are similar to rock fall distributions noted elsewhere (Guerin et al. 2014), and 277
increase progressively from the first year, reflective of greater proportions of smaller events (or 278
lower proportions of higher magnitudes) recorded within the size-frequency domain. Without 279
continuous monitoring data it is not possible to correlate the occurrence of failures with specific 280
triggers. However, the differences in spatial distributions of the failures recorded may reflect 281
13
the relative quiescence in storm incidence and a reduction in the ice season following the 2014 282
– 2015 survey period, resulting in an increasing relative significance of smaller sized events. 283
The length of the ice building season, defined by the number of frost days (simplified to 284
temperatures < 0 °C), was 34 days longer during the first survey period (calculated from 1st July 285
2014 to 30th June 2015) than in the following two years (2015 – 2017, divided in the same 286
manner). Frost days have been found to be the only significant climatic influence on retreat rate 287
in a global analysis of rock coast erosion rates (Prémaillon et al. 2018), although direct causal 288
relationships are complicated by lagged responses. 289
290
All annual rock fall distributions exhibit an inversion or roll-over at small magnitudes (< 0.001 291
m3), but not at the highest magnitudes detected (up to 10 m3). This under-representation of low 292
magnitude changes in all survey data corresponds approximately to the minimum detection 293
threshold applied to the data. Roll-over effects have been well documented in power-law 294
distributions elsewhere (Barlow et al. 2012), but questions remain over whether they reflect 295
genuine process effects or under-sampling resulting from the sensitivity of the change detection 296
(van Veen et al. 2017). 297
298
14
299
Figure 2. Annual change map of the Arctic rock cliffs at Veslebogen; SfM photogrammetry has 300
been used to remove secondary surface changes associated with material reworking and beach 301
level fluctuations. Magnitude-frequency diagrams below show a greater proportion smaller 302
failures occurred during the 2015-2016 and 2016-2017 survey periods relative to that in 2014-303
2015 for the same monitored area. All three of the volume-frequency domains contain roll-over 304
effects at lower magnitudes (denoted by the red squares). 305
Cullen et al (2018) conducted a comprehensive analysis that demonstrated the ability of close-306
range SfM to detect sub-millimetre change on rock surfaces of low relative topographic 307
roughness. Many other studies have successfully employed TMEM data to assess surface 308
downwearing processes (Strzelecki et al., 2017), but the spatial distribution and qualitative 309
information gained from the SfM photogrammetry provide clear advantages to understanding 310
micro-scale (>1 x 10-6 m3) surface change whilst maintaining a relatively close agreement with 311
the TMEM dial gauge physical readings (Fig. 3; Cullen et al. 2018). In accordance with Cullen 312
et al. (2018) we find relative topographic complexity a limitation on the accuracy of the high 313
resolution, close range data photogrammetric data. The bay sub-sections had a slightly higher 314
15
recession rate than the spurs (m2 rates of 0.0006 m and 0.0004 m per yr respectively). When 315
multiplied up to the scale of the whole cliff, rather than a roll-over below 1 x 10-3 m3, high 316
frequencies of low magnitude events appear to diverge from the power law at much smaller 317
scales (below 1 x 10-6 m3; Fig. 3). The exact nature of the inflection, that appears to be scalable 318
and relatively consistent across the several orders of magnitude detectable by the high 319
resolution photogrammetric approach, remains undetermined, falling between the survey scales 320
used. It is evident that spatial variability may play a key role in the low magnitude end of the 321
size-frequency domain (also noted in Lim et al. 2010), placing greater emphasis on the 322
identification of geomorphic process-responses spatially across the cliff. 323
324
325
326
Fig. 3. Volumetric magnitude-frequency losses during three years of monitoring using TLS and 327
photogrammetry for changes for monitored (solid lines) and modelled (dotted and dashed lines) 328
magnitudes exceeding 1 x 10-3 m3 and close range photogrammetry for changes below 1 x 10-6 329
m3 (a minimum threshold of 1 x 10-8 has been imposed to reflect the inability to use TMEM 330
16
validation on the rock wall). The image (top right) shows a SfM difference map (2015 to 2016) 331
above a comparison between the average rate of surface lowering detected by TMEM and SfM 332
on the shore platform to validate the approach. It should be noted that the TMEM detected 333
swelling (positive elevation change), a genuine process in shore platforms noted by Stephenson 334
and Kirk (2001) in multiple readings that reduce the averaged surface lowering result. 335
336
4.2 Arctic rock coast geomorphic zones 337
Thermistor strings drilled into the cliff and platform either side of the Wilczekodden rock 338
peninsula (Fig. 1b) demonstrate the interplay between cryogenic and marine influence. The 339
moderating effects of coastal waters in summer and ice-foot in winter are evident in the muted 340
rock platform thermal variations relative to those undergone by the cliff rock material (Fig. 4). 341
The implication is that the efficacy of thermal stress effects (Collins and Stock 2016) will be 342
orders of magnitude greater horizontally into cliff material than vertically through platform 343
material, forming a potential link between cryogenic temperature variations and strandflat 344
enhancement. In early January 2016 there was a warming event related to an Atlantic wind 345
storm designated ‘Frank’ by the U.K. Met Office that dragged warm moist air north into the 346
Arctic (Kim et al. 2017). All rock to a depth of 0.5 m within both the cliff and platform recorded 347
a distinguishable positive thermal response to these anomalous conditions (Fig. 4bi). The most 348
pronounced warming was recorded in the west-facing cliffs (2 °C warmer than the east facing 349
cliffs), which are more exposed to oceanic and storm influence. The influence of local setting 350
controls is also evident in a rapid cooling event that occurred only in the west shore platform in 351
late April (Fig. 4bii). Time-lapse imagery shows that this cooling was a result of storm driven 352
break-up of the ice-foot, as cold water ingress reduced platform rock temperatures by several 353
degrees within three hours (Fig. 4c). The more sheltered east-facing cliffs were insulated by an 354
ice-foot until late May. Subsequent short-term (several days) cooling effects from occasional 355
17
iceberg rafts collecting on the east-facing coast have also been recorded from late May to early 356
June and then less frequently throughout the summer. These influences of winter storm 357
exposure and ice-foot dynamics on west-facing coasts and summer iceberg rafts from glacial 358
discharge on the east-facing rock coast lead to highly responsive, locally conditioned seasonal 359
thermal regimes (Fig. 4a). 360
361
Thermal links to rock fracture mechanics have been demonstrated as potential drivers of cyclic 362
patterns of cumulative damage and ultimately failure (Collins and Stock 2016). The penetration 363
of thermal variance within the Arctic rock cliffs declines by over 60% below 0.1 m (to 0.3 m) 364
on both the west and east coasts, a depth which also corresponds with the mean rock fall depth 365
recorded during that period. We infer that thermal stress may provide a control on rock fall 366
development and occurrence under Arctic conditions, similar to that established for exfoliation 367
sheets (Collins and Stock 2016), although other environmental factors such as bioprotection 368
(Coombes et al. 2017) and humidity (Eppes and Keanini 2017) may also condition the 369
geomorphic response. 370
371
18
372
Fig. 4. Temperature change with depth in Arctic rock cliffs and platforms (a), Westerly and 373
easterly aspects of the Wilczekodden peninsula (Fig. 1) are exposed to oceanic storm and 374
iceberg influence respectively. Summary cliff face statistics are provided for both summer and 375
winter periods (b), Rock temperatures through the year in shore platforms demonstrate a 376
muted response relative to cliff face variations. Key thermal signatures highlighted (i and ii) 377
represent an anomalous warming period and storm driven cooling event as break-up of the 378
insulating ice-foot allowed cold water ingress over the platform (c), Before and after ice-foot 379
break-up images. 380
381
The three-dimensional temperature map of the Veslbogen cliffs revealed a general decrease in 382
temperature from the base of the cliff upwards (Fig. 5). However, the temperature patterns were 383
not even and in the sheltered embayments the weathered back walls demonstrated lower 384
temperatures down to the cliff base, likely due to higher moisture content. The protruding areas 385
19
displayed clearer zonation that appears to reflect the abrasive action at the cliff base 386
transitioning up in layers and temperature increments to a 1.5 m thick capping layer of heavily 387
weathered, compacted material above the rock mass. 388
389
The results of the rock hardness profiles show a quantitative transition, decreasing with height 390
up the cliff (Fig. 5). Here we suggest that the thermal and hardness differences reflect 391
geomorphic process zones. Where the cliff base is impacted by dynamic ice contact from 392
iceberg rafts, from ice-foot formation and movement, or from storm waves, the erosive agency 393
is particularly effective. This agency is reflected by high rates of back wearing and by peak 394
rock hardness (Blanco-Chao et al. 2007; Strzelecki et al., 2017). The ice-contact zone extends 395
above the high water mark and on the most exposed spurs protruding out from the general line 396
of the cliff face the scour and abrasive processes extend up to two thirds of the cliff height (‘B’ 397
In Fig. 5b). In more sheltered areas such as embayments where ice movement potential is 398
limited, this zone does not appear to exist and a weaker weathering zone dominates down to the 399
cliff foot (‘A’ In Fig. 5b). The hardness values in these areas were up to a third lower than those 400
recorded at the scoured cliff base. There is a transition zone between the scour and weathered 401
zones where intermediate hardness values have been recorded. Questions remain over the origin 402
of this transition, potentially evidencing historic scour processes that no longer, or rarely, persist 403
at these elevations (or indeed removed under high rates of isostatic uplift) or the influence of 404
extreme storm events that occur at a frequency of several years (Degard and Sollid 1993). These 405
process zones have been visualised for the first time using three-dimensional temperature 406
mapping, although further work is required to fully explore the potential of this approach for 407
geomorphic studies. 408
409
410
20
411
412
Fig. 5. Photogrammetric cliff surface model (a) and coincident three-dimensional temperature 413
point cloud (b, coloured to reflect the inferred severity of weathering), with a vertical profile of 414
Schmidt hammer readings (25 values averaged to a mean rebound value, R) over the accessible 415
section of the monitored cliff. The lower image (c) is the site in March (2017) when the ice-416
foot transitions to decay; there appears to be a good spatial agreement with the thermal zones 417
noted during the summer survey data. 418
419
The rock coast environment at Hornsund has been subjected to some of the most dramatic 420
temperature shifts in the Arctic over recent decades (Hinzman et al. 2005), and as such its wider 421
significance is twofold. Firstly, it provides an opportunity to explore geomorphic behaviour 422
under extreme shifts in process activity. Models of rock coast development and response have 423
often been linked to sea-level (Trenhaile 2010), significant wave height (Norman et al. 2013) 424
and seasonality (Hansom 2014) but quantitative links between process environment changes 425
21
and rock cliff material responses have remained controversial and challenging (Naylor et al. 426
2010; Rosser et al. 2013). Secondly, the sustained temperature rise experienced at Hornsund 427
may serve as an indicator of changes likely to impact other regions as climate trends continue. 428
The relative difference in rock hardness across targeted elevations within rock coast profiles 429
provides a potentially simple quantitative indicator of the presence or absence and significance 430
of geomorphic zones at other sites (Strzelecki et al., 2017). 431
A regional north-south transect of data on temperature variations and rock hardness profiles has 432
been compiled (Fig. 6). Average hardness profiles have been taken through rock cliffs from 433
representative lithologies at each site, but for comparability and clarity lithologies of similar 434
hardness values have been presented. The hardness recorded within specific rock coast zones 435
at any particular site are influenced by the local competence of the cliff material, but by using 436
the same approach, relative differences in intact rock hardness can be compared using the 437
standardised rebound value (Hansen et al. 2013). The hardness values essentially provide a 438
chronology of weathering exposure (Kellerer-Pirklbauer et al. 2008), assuming no armouring 439
effects of precipitated solutes or biofilms that may artificially increase hardness, processes 440
noted in arid extreme environments (Viles and Goudie 2004). 441
The west coast of Spitsbergen has undergone variable levels of warming since 1957, with peak 442
temperature increases occurring in the south (Hornsund) and central inner fjord areas 443
(Longyearbyen). In such areas, the warming has been most pronounced in the winter season 444
when ice foot development occurs, producing rock hardness transitions from a strong peak in 445
the intertidal and sea-ice zone to rapidly declining through the depleted ice-foot and subaerially 446
weathered zones. By contrast, areas that have undergone less severe temperature increases 447
maintain the average rock hardness level upwards through the sea-ice and ice-foot zone. For 448
example, the mean hardness rebound values (R) for both the scoured, fresh intertidal rock and 449
the subaerially weathered zone higher up the cliff face are comparable for Billefjorden and 450
22
Skilvika (differences of R=1 and R=2 respectively between the two sites in the intertidal and 451
weathered zones), but there is a notable difference (R=6) in the hardness of the ice-foot zone 452
measurements (Fig. 6). Therefore, hardness profiles can effectively identify weathering process 453
zones without the need to make inferences on the driving processes, which can be intrinsic or 454
extrinsic and often lack significance in global scale comparisons (Prémaillon et al. 2018). 455
456
457
Fig. 6. Mean monthly temperatures (1957 and 2016) and the resultant differences at sites in a 458
north-south transect through western Spitsbergen (left) and the averaged rock hardness values 459
of key geomorphic process zones superimposed on a schematic diagram of rock coast 460
morphology (right). 461
462
5. Implications 463
23
Arctic coastal rock cliff retreat is commonly thought to be amongst the most effective of all 464
denudation environments (Prick 2003). Cryogenic weathering and frost wedging in particular 465
have been found to play a significant role in the behaviour and material failure of rock walls 466
(Degard and Sollid 1993; Matsuoka 2008), but recent concerns have focussed on the 467
climatically driven increasing exposure at the coast (Sessford et al. 2015). Increases in the open 468
water season and in the intensity of storms associated with rising temperature has been widely 469
hypothesised to result in an increase in the rates of erosion on Arctic coastlines (Hinzman et al. 470
2005; Wojtysiak et al. 2018). Whilst valid for highly erosive coastal systems such as ice rich 471
permafrost cliffs (Jones et al. 2009), the response of rock coasts appears to be more complex. 472
Historic rates of rock cliff change are extremely sparse and although the detailed study 473
conducted by Jahn (1961) suggests that rates of rock wall recession have declined over this 474
period there is insufficient validation to make a reliable quantitative assessment. However, the 475
qualitative information is just as revealing. In 1958 the ice-foot is recorded to have persisted 476
almost throughout the summer season (Jahn 1961). During the 2014 – 2017 surveys presented 477
here, the ice-foot formed during early November and was fully developed by late January, but 478
the key difference with 1958 conditions appears to be that the ice-foot was still building 479
throughout April (based on historic temperatures) whereas it now degrades through April and 480
is gone by late May. The thermal regime during ice-foot formation is critical to its longevity 481
through the melt season and the incorporation of icebergs from glacier outlets may provide a 482
key control in ice-foot morphology, resulting in rougher, more sporadic accumulations that 483
reach higher up the shore and cliff face (Wiseman et al. 1981). The Hansbreen glacier generates 484
drifts of icebergs that reach the Veslebogen coastline following storm events that carry them 485
out of the fjord. In 1957 the Hansbreen terminus reached the valley mouth and so calving events 486
were released directly into the open coast providing a significantly greater frequency of icebergs 487
to the cliff base (Vieli et al. 2002). Therefore, the impact of warming over the intervening period 488
24
is twofold, producing both a reduction in the ice-foot formation duration and intensity and a 489
reduction in the supply of ice from retreating and increasingly disconnected glacier systems. 490
There is no indication that the temperature shifts in the Arctic, which have been particularly 491
pronounced on the west coast of Spitsbergen, have transitioned from ice foot formation under 492
storm and drift driven processes to tidal freezing processes that exert much more regular and 493
height-constrained influence up the shoreface (Wiseman et al. 1981). There is potential to 494
explore these system level responses through quantitative monitoring of rock hardness, perhaps 495
through the ice free season and over successive seasons, particularly when combined with 496
remotely sensed relative indicators of coastal process zones such as the three-dimensional 497
thermal mapping presented here. The results the from historical study comparison suggest that 498
the decline and ultimate loss of ice-foot processes (largely mechanical such as plucking and 499
abrasive scour) from Arctic rock cliffs may currently be a more significant factor than the 500
increased exposure to higher energy storms; dominating the geomorphic response. Further to 501
this, the increased intensity of Arctic storms in spring noted by Wojtysiak et al. (2018), may 502
expedite the break-up and removal of the ice-foot relative to past records of when it persisted 503
year-long (Wiseman et al. 1981). It is increasingly acknowledged that the temperature driven 504
changes to the open water season affecting Svalbard is producing complex and variable coastal 505
erosion responses that require further investigation (Sessford et al. 2015). 506
The distinctiveness and effectiveness of coastal Arctic cliff processes has been found to set 507
them apart from other Arctic rock wall environments (Wangensteen et al. 2007). The fit of the 508
size-frequency domains to power-law distributions holds significant potential in understanding 509
the relation between driving processes and responding rates and mechanisms of change and in 510
predicting future developments (Teixeira 2006). The power law distributions break down at 511
changes below 1 x 10-3 m3 (and there are possible indications of divergence from this scaling 512
in magnitudes up to 1 x 10-2 m3), with lower than modelled numbers being detected. This is a 513
25
common occurrence in remotely sensed datasets of change and could be due to either 514
incompleteness in the record or genuine mechanistic controls (Guerin et al. 2014; Guzzetti et 515
al. 2002). Here, to complete a rigorous assessment of the data, subsections of close range 516
photogrammetric monitoring indicate that the power law scaling may hold to finer scales but at 517
a reduced gradient and less variation between the scales of change occurring. These data have 518
implications for understanding the geomorphic significance of the small scale changes that are 519
often ignored in rock slope studies and highlight the impact of thresholding approaches 520
commonly used to isolate and quantify volumetric change in survey datasets. However, the 521
representation of such small areas relative to the wider behaviour of the rock wall remains open 522
to question, particularly given the potential for spatially varied responses to the cryogenic, 523
marine and subaerial processes in operation (Naylor et al. 2012; Strzelecki 2017; Wangensteen 524
et al. 2007). 525
Many Arctic coastal systems are experiencing altered thermal and hydrological regimes 526
associated with transitions from glacial to deglacial conditions. The use of thermal mapping 527
over landforms appears to hold potential for detecting moisture and temperature related 528
weathering zones and thus in characterising geomorphic responses that are not readily visible 529
with colour imagery. It has particular utility in the detection of moisture related influences, 530
critical to periglacial environments (Hall et al. 2002). The focus here has been on the relative 531
differences in temperature to aid characterisation and classification, but the close (±1 °C) 532
agreement with point based checks and the supporting differences in rock hardness readings 533
also highlight the potential quantitative value of the approach. 534
The ability to delineate zones of geomorphic response provides the basis to more focused and 535
rigorous investigations of behaviour, refining and contextualising rate estimates spatially and 536
helping to inform rock coast models (Naylor et al. 2012). Questions raised from this research 537
concern the origin and future evolution of these zones. In particular, whether the transitionary 538
26
semi-weathered zone identified above the scoured base corresponds with storm waves and may 539
increase in prominence with higher intensity events or represents an inherited feature from 540
previous decades when the ice-foot extent remained higher for longer and will consequently 541
decline in significance. The influence of ice on coastal geomorphology has the potential to both 542
enhance erosion through scour, plucking and abrasion (Are et al. 2008; Dionne J.-C. 1988) and 543
to mitigate it through thermal insulation (Scrosati and Eckersley 2007) and by attenuating, 544
dissipating and removing wave and tidal influence from the cliff face (Forbes et al., 1986). The 545
spatial and temporal variability of ice-foot processes has long been recognised (Wiseman et al. 546
1981), complicating the interpretation and analysis of climatic influences on Arctic rock coast 547
geomorphic behaviour. However, the indications of abrasive geomorphic zones that extend 548
significantly beyond storm wave height suggest that cryogenic control may be a dominant but 549
declining influence on these systems. These understudied aspects of Arctic coast landforms, 550
often typified by either very low (André 1997) or very high (Overduin 2014) rates and 551
intensities of weathering, erosion and denudation processes, have important implications for 552
primary ecosystem development and nutrient availability (Borin et al. 2010) and for asset 553
management of coastal infrastructure (Ford et al. 2010). 554
6. Conclusion 555
The Arctic is undergoing system-wide responses to climate-induced change (Hinzman et al. 556
2005), affording a rare opportunity to assess rock coast responses in the context of process 557
conditions that are altering over relatively recent (>100 year) timescales. Utilising high 558
resolution survey approaches, novel temperature mapping and monitoring and simple 559
geotechnical characterisation we draw the following conclusions: 560
27
A reduction in sea-ice and greater storm intensities recorded in the Arctic have the 561
potential to accelerate erosion rates (Wojtysiak et al. 2018), but here we find rates of 562
rock coast erosion are lower than those reported at the same site in 1961 (Jahn 1961). 563
Marked increases in temperatures during the winter months critical to ice-foot 564
formation have resulted in significant decreases to the extent and duration of the 565
present-day ice-foot. 566
Power law relationships effectively model Arctic rock cliff behaviour over several 567
orders of magnitude, but an inflection demonstrated through lower exponents was 568
detected in the scaling of low magnitude (below 1 x 10-6 m3) rock falls. 569
Temperature monitoring with depth into the shore platform, cliff face and cliff top has 570
revealed the sensitivity of Arctic rock coasts to both global teleconnection events such 571
as severe depressions and local influences such as ice-foot dynamics and the presence 572
of icebergs. 573
The addition here of three-dimensional thermal mapping to spatially map process 574
zones for the first time, validated through quantitative rock hardness measurements, 575
suggests a declining cryogenic influence in the most temperature affected areas of 576
Spitsbergen. This approach provides new opportunities to monitor, analyse, interpret 577
and predict Arctic coastal rock cliff responses in the context of altered, climate-driven 578
process environments. 579
580
Acknowledgements 581
The authors thank and acknowledge the National Science Centre, Poland, for funding 582
‘POROCO – Mechanisms controlling the evolution and geomorphology of rock coasts in polar 583
28
climates’ (UMO2013/11/B/ST10/00283). M.C.S. has been also supported by the Foundation 584
for Polish Science and the Ministry of Science and Higher Education in Poland. The authors 585
thank the staff of Polish Polar Station in Hornsund, Svalbard, for support in field logistics. We 586
gratefully acknowledge the insightful and helpful reviewers comments used to improve the 587
paper. 588
589
References 590
André, M.‐F. (1997). Holocene Rockwall Retreat in Svalbard: A Triple‐Rate Evolution. Earth Surface 591 Processes and Landforms, 22, 423‐440 592 Are, F., Reimnitz, E., Grigoriev, M., Hubberten, H.W., & Rachold, V. (2008). The influence of cryogenic 593 processes on the erosional arctic shoreface. Journal of Coastal Research, 24, 110‐121 594 Barlow, J., Lim, M., Rosser, N., Petley, D., Brain, M., Norman, E., & Geer, M. (2012). Modeling cliff 595 erosion using negative power law scaling of rockfalls. Geomorphology, 139, 416‐424 596 Barnes, D.K.A. (1999). The influence of ice on polar nearshore benthos. Journal of the Marine 597 Biological Association of the United Kingdom, 79, 401‐407 598 Berthling, I., & Etzelmuller, B. (2011). The concept of cryo‐conditioning in landscape evolution. 599 Quaternary Research, 75, 378‐384 600 Blanco‐Chao, R., Pérez‐Alberti, A., Trenhaile, A.S., Costa‐Casais, M., & Valcárcel‐Díaz, M. (2007). 601 Shore platform abrasion in a para‐periglacial environment, Galicia, northwestern Spain. 602 Geomorphology, 83, 136‐151 603 Borin, S., Ventura, S., Tambone, F., Mapelli, F., Schubotz, F., Brusetti, L., Scaglia, B., D'Acqui, L.P., 604 Solheim, B., Turicchia, S., Marasco, R., Hinrichs, K.‐U., Baldi, F., Adani, F., & Daffonchio, D. (2010). 605 Rock weathering creates oases of life in a High Arctic desert. Environmental Microbiology, 12, 293‐606 303 607 Budetta, P., Galietta, G., & Santo, A. (2000). A methodology for the study of the relation between 608 coastal cliff erosion and the mechanical strength of soils and rock masses. 609 Byrne, M.L., & Dionne, J.C. (2002). Typical aspects of cold regions shorelines. 610 Collins, B.D., & Stock, G.M. (2016). Rockfall triggering by cyclic thermal stressing of exfoliation 611 fractures. Nature Geoscience, 9, 395‐+ 612 Coombes, M.A. (2011). Rock warming and drying under simulated intertidal conditions, part I: 613 experimental procedures and comparisons with field data. Earth Surface Processes and Landforms, 614 36, 2114‐2121 615 Coombes, M.A., Viles, H.A., Naylor, L.A., & La Marca, E.C. (2017). Cool barnacles: Do common 616 biogenic structures enhance or retard rates of deterioration of intertidal rocks and concrete? Science 617 of the Total Environment, 580, 1034‐1045 618 Cullen, N.D., Verma, A.K., & Bourke, M.C. (2018). A comparison of structure from motion 619 photogrammetry and the traversing micro‐erosion meter for measuring erosion on shore platforms. 620 Earth Surf. Dynam., 6, 1023‐1039 621 Degard, R.S., & Sollid, J.L. (1993). Coastal cliff temperatures related to the potential for cryogenic 622 weathering processes, western Spitsbergen, Svalbard. Polar Research, 12, 95‐106 623 Dionne J.‐C., B.D. (1988). Frost weathering and ice action in shore platform development, with 624 particular reference to Quebec, Canada. Zeitschrift Fur Geomorphologie, Supplement band 71, 117‐625 130 626 Edis, E., Flores‐Colen, I., & de Brito, J. (2014). Passive thermographic detection of moisture problems 627 in façades with adhered ceramic cladding. Construction and Building Materials, 51, 187‐197 628
29
Eppes, M.‐C., & Keanini, R. (2017). Mechanical weathering and rock erosion by climate‐dependent 629 subcritical cracking. Reviews of Geophysics, 55, 470‐508 630 Etzelmüller, B., Schuler, T., Isaksen, K., Christiansen, H., Farbrot, H., & Benestad, R. (2011). Modeling 631 the temperature evolution of Svalbard permafrost during the 20th and 21st century. 632 Feagin, R.A., Williams, A.M., Popescu, S., Stukey, J., & Washington‐Allen, R.A. (2014). The Use of 633 Terrestrial Laser Scanning (TLS) in Dune Ecosystems: The Lessons Learned. Journal of Coastal 634 Research, 30, 111‐119 635 Forbes, D.L. (2011). State of the Arctic Coast 2010 – Scientific Review and Outlook. In (p. 178). 636 Helmholtz‐Zentrum, Geesthacht, Germany: International Permafrost Association 637 Forbes, D.L., & Taylor, R.B. (1994). ICE IN THE SHORE ZONE AND THE GEOMORPHOLOGY OF COLD 638 COASTS. Progress in Physical Geography, 18, 59‐96 639 Ford, J.D., Bell, T., & St‐Hilaire‐Gravel, D. (2010). Vulnerability of Community Infrastructure to Climate 640 Change in Nunavut: A Case Study From Arctic Bay. In G.K. Hovelsrud, & B. Smit (Eds.), Community 641 Adaptation and Vulnerability in Arctic Regions (pp. 107‐130). Dordrecht: Springer Netherlands 642 Fritz, M., Vonk, J.E., & Lantuit, H. (2017). COMMENTARY: Collapsing Arctic coastlines. Nature Climate 643 Change, 7, 6‐7 644 Gilham, J., Barlow, J., & Moore, R. (2018). Marine control over negative power law scaling of mass 645 wasting events in chalk sea cliffs with implications for future recession under the UKCP09 medium 646 emission scenario. Earth Surface Processes and Landforms, 43, 2136‐2146 647 Gjelten, H.M., Nordli, O., Isaksen, K., Forland, E.J., Sviashchennikov, P.N., Wyszynski, P., Prokhorova, 648 U.V., Przybylak, R., Ivanov, B.V., & Urazgildeeva, A.V. (2016). Air temperature variations and 649 gradients along the coast and fjords of western Spitsbergen. Polar Research, 35 650 Grinzato, E., Vavilov, V., & Kauppinen, T. (1998). Quantitative infrared thermography in buildings. 651 Energy and Buildings, 29, 1‐9 652 Guerin, A., Hantz, D., Rossetti, J.P., & Jaboyedoff, M. (2014). Brief communication"Estimating rockfall 653 frequency in a mountain limestone cliff using terrestrial laser scanner". Nat. Hazards Earth Syst. Sci. 654 Discuss., 2014, 123‐135 655 Guzzetti, F., Malamud, B., L Turcotte, D., & Paola, R. (2002). Power‐law correlations of landslide areas 656 in central Italy. 657 Hall, K. (1988). A laboatory simulation of rock breakdown due to freeze thaw in a maritime Antarctic 658 environment. Earth Surface Processes and Landforms, 13, 369‐382 659 Hall, K., Thorn, C.E., Matsuoka, N., & Prick, A. (2002). Weathering in cold regions: some thoughts and 660 perspectives. Progress in Physical Geography, 26, 577‐603 661 Hansen, C.D., Meiklejohn, K.I., Nel, W., Loubser, M.J., & Van der merwe, B.J. (2013). Aspect‐662 controlled weathering observed on a blockfield in dronning maud land, antarctica. Geografiska 663 Annaler: Series A, Physical Geography, 95, 305‐313 664 Hansom, J.D. (2014). The rock coasts of polar and sub‐polar regions. In K. D.M. (Ed.), Rock Coast 665 Geomorphology: A Global Synthesis. (pp. 263–504): Geological Society, London, Memoirs 666 Herman, A., Wojtysiak, K., & Moskalik, M. (2019). Wind wave variability in Hornsund fjord, west 667 Spitsbergen. Estuarine, Coastal and Shelf Science, 217, 96‐109 668 Hinzman, L.D., Bettez, N.D., Bolton, W.R., Chapin, F.S., Dyurgerov, M.B., Fastie, C.L., Griffith, B., 669 Hollister, R.D., Hope, A., Huntington, H.P., Jensen, A.M., Jia, G.J., Jorgenson, T., Kane, D.L., Klein, D.R., 670 Kofinas, G., Lynch, A.H., Lloyd, A.H., McGuire, A.D., Nelson, F.E., Oechel, W.C., Osterkamp, T.E., 671 Racine, C.H., Romanovsky, V.E., Stone, R.S., Stow, D.A., Sturm, M., Tweedie, C.E., Vourlitis, G.L., 672 Walker, M.D., Walker, D.A., Webber, P.J., Welker, J.M., Winker, K., & Yoshikawa, K. (2005). Evidence 673 and implications of recent climate change in northern Alaska and other arctic regions. Climatic 674 Change, 72, 251‐298 675 Inkpen, R.J., Stephenson, W.J., Kirk, R.M., Hemmingsen, M.A., & Hemmingsen, S.A. (2010). Analysis of 676 relationships between micro‐topography and short‐ and long‐term erosion rates on shore platforms 677 at Kaikoura Peninsula, South Island, New Zealand. Geomorphology, 121, 266‐273 678 Jahn, A. (1961). Quantitative analysis of some periglacial processes in Spitsbergen. [Warszawa]: 679 [Panstwowe Wydawn. Naukowe] 680
30
Jakacki, J., Przyborska, A., Kosecki, S., Sundfjord, A., & Albretsen, J. (2017). Modelling of the Svalbard 681 fjord Hornsund. Oceanologia, 59, 473‐495 682 Jones, B.M., Arp, C.D., Jorgenson, M.T., Hinkel, K.M., Schmutz, J.A., & Flint, P.L. (2009). Increase in 683 the rate and uniformity of coastline erosion in Arctic Alaska. Geophysical Research Letters, 36 684 Kellerer‐Pirklbauer, A., Wangensteen, B., Farbrot, H., & Etzelmuller, B. (2008). Relative surface age‐685 dating of rock glacier systems near Holar in Hjaltadalur, northern Iceland. Journal of Quaternary 686 Science, 23, 137‐151 687 Kim, B.M., Hong, J.Y., Jun, S.Y., Zhang, X.D., Kwon, H., Kim, S.J., Kim, J.H., Kim, S.W., & Kim, H.K. 688 (2017). Major cause of unprecedented Arctic warming in January 2016: Critical role of an Atlantic 689 windstorm. Scientific Reports, 7 690 Kirk, R.M. (1977). Rates and forms of erosion on intertidal platforms at Kaikoura Peninsula, South 691 Island, New Zealand. New Zealand Journal of Geology and Geophysics, 20, 571‐613 692 Krautblatter, M., Moser, M., Schrott, L., Wolf, J., & Morche, D. (2012). Significance of rockfall 693 magnitude and carbonate dissolution for rock slope erosion and geomorphic work on Alpine 694 limestone cliffs (Reintal, German Alps). Geomorphology, 167‐168, 21‐34 695 Lantuit, H., Overduin, P.P., Couture, N., Wetterich, S., Are, F., Atkinson, D., Brown, J., Cherkashov, G., 696 Drozdov, D., Forbes, D.L., Graves‐Gaylord, A., Grigoriev, M., Hubberten, H.W., Jordan, J., Jorgenson, 697 T., Odegard, R.S., Ogorodov, S., Pollard, W.H., Rachold, V., Sedenko, S., Solomon, S., Steenhuisen, F., 698 Streletskaya, I., & Vasiliev, A. (2012). The Arctic Coastal Dynamics Database: A New Classification 699 Scheme and Statistics on Arctic Permafrost Coastlines. Estuaries and Coasts, 35, 383‐400 700 Lim, M., Petley, D.N., Rosser, N.J., Allison, R.J., Long, A.J., & Pybus, D. (2005). Combined digital 701 photogrammetry and time‐of‐flight laser scanning for monitoring cliff evolution. Photogrammetric 702 Record, 20, 109‐+ 703 Lim, M., Rosser, N.J., Allison, R.J., & Petley, D.N. (2010). Erosional processes in the hard rock coastal 704 cliffs at Staithes, North Yorkshire. Geomorphology, 114, 12‐21 705 Lim, M., Rosser, N.J., Petley, D.N., & Keen, M. (2011). Quantifying the Controls and Influence of Tide 706 and Wave Impacts on Coastal Rock Cliff Erosion. Journal of Coastal Research, 27, 46‐56 707 Limber, P.W., & Murray, A.B. (2011). Beach and sea‐cliff dynamics as a driver of long‐term rocky 708 coastline evolution and stability. Geology, 39, 1147‐1150 709 Luscombe, D.J., Anderson, K., Gatis, N., Grand‐Clement, E., & Brazier, R.E. (2015). Using airborne 710 thermal imaging data to measure near‐surface hydrology in upland ecosystems. Hydrological 711 Processes, 29, 1656‐1668 712 Matsuoka, N. (2008). Frost weathering and rockwall erosion in the southeastern Swiss Alps: Long‐713 term (1994–2006) observations. 714 Matsuoka, N., Moriwaki, K., & Hirakawa, K. (1996). Field experiments on physical weathering and 715 wind erosion in an Antarctic cold desert. Earth Surface Processes and Landforms, 21, 687‐699 716 Naylor, L.A., Coombes, M.A., & Viles, H.A. (2012). Reconceptualising the role of organisms in the 717 erosion of rock coasts: A new model. Geomorphology, 157‐158, 17‐30 718 Naylor, L.A., Spencer, T., Lane, S.N., Darby, S.E., Magilligan, F.J., Macklin, M.G., & Möller, I. (2017). 719 Stormy geomorphology: geomorphic contributions in an age of climate extremes. Earth Surface 720 Processes and Landforms, 42, 166‐190 721 Naylor, L.A., Stephenson, W.J., & Trenhaile, A.S. (2010). Rock coast geomorphology: Recent advances 722 and future research directions. Geomorphology, 114, 3‐11 723 Norman, E.C., Rosser, N.J., Brain, M.J., Petley, D.N., & Lim, M. (2013). Coastal cliff‐top ground 724 motions as proxies for environmental processes. Journal of Geophysical Research‐Oceans, 118, 6807‐725 6823 726 Overduin, P.P. (2014). Coastal changes in the Arctic. In I.P.W. Martini, H.R. (Ed.), Sedimentary coastal 727 zones from high to low latitudes: similarities and differences. (pp. 103‐129): Geological Society, 728 London, Special Publications 729 Overeem, I., Anderson, R.S., Wobus, C.W., Clow, G.D., Urban, F.E., & Matell, N. (2011). Sea ice loss 730 enhances wave action at the Arctic coast. Geophysical Research Letters, 38 731
31
Parkinson, C.L., & Comiso, J.C. (2013). On the 2012 record low Arctic sea ice cover: Combined impact 732 of preconditioning and an August storm. Geophysical Research Letters, 40, 1356‐1361 733 Prémaillon, M., Regard, V., Dewez, T.J.B., & Auda, Y. (2018). GlobR2C2 (Global Recession Rates of 734 Coastal Cliffs): a global relational database to investigate coastal rocky cliff erosion rate variations. 735 Earth Surf. Dynam., 6, 651‐668 736 Prick, A. (2002). Monitoring weathering and erosion of bedrock on a coastal cliff, Longyearbyen, 737 Svalbard. In V.B. Rachold, J; Solomon, S; Sollid, J.L. (Ed.), Arctic coastal dynamics: report of the 3rd 738 International Workshop University of Oslo (Norway) 2‐5 December 2002. (pp. 95‐95). Oslo, Norway: 739 Reports on Polar and Marine Research 740 Prick, A. (2003). Frost weathering and rock fall in an arctic environment, Longyearbyen, Svalbard. 741 Rapp, A. (1957). Studien über Schutthalden in Lappland und auf Spitzbergen. . Zeitschrift für 742 Geomorphologie, 1, 179‐200 743 Rapp, A. (1960). Recent Development of Mountain Slopes in Kärkevagge and Surroundings, Northern 744 Scandinavia. Geografiska Annaler, 42, 65‐200 745 Rasmussen, T.L., Thomsen, E., Slubowska, M.A., Jessen, S., Solheim, A., & Koc, N. (2007). 746 Paleoceanographic evolution of the SW Svalbard margin (76 degrees N) since 20,000 C‐14 yr BP. 747 Quaternary Research, 67, 100‐114 748 Rohmer, J., & Dewez, T. (2015). Analysing the spatial patterns of erosion scars using point process 749 theory at the coastal chalk cliff of Mesnil‐Val, Normandy, northern France. Natural Hazards and Earth 750 System Sciences, 15, 349‐362 751 Rosser, N., Lim, M., Petley, D., Dunning, S., & Allison, R. (2007). Patterns of precursory rockfall prior 752 to slope failure. Journal of Geophysical Research‐Earth Surface, 112 753 Rosser, N.J., Brain, M.J., Petley, D.N., Lim, M., & Norman, E.C. (2013). Coastline retreat via 754 progressive failure of rocky coastal cliffs. Geology, 41, 939‐942 755 Sass, O. (2005). Rock moisture measurements: techniques, results, and implications for weathering. 756 Earth Surface Processes and Landforms, 30, 359‐374 757 Scrosati, R., & Eckersley, L.K. (2007). Thermal insulation of the intertidal zone by the ice foot. Journal 758 of Sea Research, 58, 331‐334 759 Sessford, E.G., Baeverford, M.G., & Hormes, A. (2015). Terrestrial processes affecting unlithified 760 coastal erosion disparities in central fjords of Svalbard. Polar Research, 34 761 Stemberk, J., Briestenský, M., & Cacoń, S. (2015). The recognition of transient compressional fault 762 slow−slip along the northern shore of Hornsund Fjord, SW Spitsbergen, Svalbard. Polish Polar 763 Research, 36, 109 ‐ 123 764 Stephenson, W.J., & Finlayson, B.L. (2009). Measuring erosion with the micro‐erosion meter—765 Contributions to understanding landform evolution. Earth‐Science Reviews, 95, 53‐62 766 Stephenson, W.J., & Kirk, R.M. (2001). Surface swelling of coastal bedrock on inter‐tidal shore 767 platforms, Kaikoura Peninsula, South Island, New Zealand. Geomorphology, 41, 5‐21 768 Strzelecki, M.C. (2011). Schmidt hammer tests across a recently deglacierized rocky coastal zone in 769 Spitsbergen ‐ is there a "coastal amplification" of rock weathering in polar climates? Polish Polar 770 Research, 32, 239‐252 771 Strzelecki, M.C. (2017). The variability and controls of rock strength along rocky coasts of central 772 Spitsbergen, High Arctic. Geomorphology, 293, 321‐330 773 Sunamura, T. (1992). Geomorphology of rocky coasts, 302 774 Teixeira, S. (2006). Slope mass movements on rocky sea‐cliffs: A power‐law distributed natural hazard 775 on the Barlavento Coast, Algarve, Portugal. 776 Trenhaile, A.S. (1983). The development of shore platforms in high latitudes. In D.E. Smith, Dawson, 777 A.G. (Ed.), Shorelines and Isostasy (pp. 77–96). London: Institute of British Geographers. 778 Trenhaile, A.S. (1997). Coastal Dynamics and Landforms. Oxford: Oxford University Press 779 Trenhaile, A.S. (2010). The effect of Holocene changes in relative sea level on the morphology of 780 rocky coasts. Geomorphology, 114, 30‐41 781
32
van Veen, M., Hutchinson, D.J., Kromer, R., Lato, M., & Edwards, T. (2017). Effects of sampling 782 interval on the frequency‐magnitude relationship of rockfalls detected from terrestrial laser scanning 783 using semi‐automated methods. Landslides, 14, 1579‐1592 784 Vieli, A., Jania, J., & Kolondra, L. (2002). The retreat of a tidewater glacier: observations and model 785 calculations on Hansbreen, Spitsbergen. Journal of Glaciology, 48, 592‐600 786 Viles, H.A., & Goudie, A.S. (2004). Biofilms and case hardening on sandstones from Al‐Quwayra, 787 Jordan. Earth Surface Processes and Landforms, 29, 1473‐1485 788 Wangensteen, B., Eiken, T., Odegard, R.S., & Sollid, J.L. (2007). Measuring coastal cliff retreat in the 789 Kongsfjorden area, Svalbard, using terrestrial photogrammetry. Polar Research, 26, 14‐21 790 Webster, C., Westoby, M., Rutter, N., & Jonas, T. (2018). Three‐dimensional thermal characterization 791 of forest canopies using UAV photogrammetry. Remote Sensing of Environment, 209, 835‐847 792 Westoby, M.J., Lim, M., Hogg, M., Pound, M.J., Dunlop, L., & Woodward, J. (2018). Cost‐effective 793 erosion monitoring of coastal cliffs. Coastal Engineering, 138, 152‐164 794 Wiseman, W.J., Owens, E.H., & Kahn, J. (1981). Temporal and spatial variability of ice‐foot 795 morphology. Geografiska Annaler Series a‐Physical Geography, 63, 69‐80 796 Wojtysiak, K., Herman, A., & Moskalik, M. (2018). Wind wave climate of west Spitsbergen: seasonal 797 variability and extreme events. Oceanologia, 60, 331‐343 798 Young, A.P., Olsen, M.J., Driscoll, N., Flick, R.E., Gutierrez, R., Guza, R.T., Johnstone, E., & Kuester, F. 799 (2010). Comparison of Airborne and Terrestrial Lidar Estimates of Seacliff Erosion in Southern 800 California. Photogrammetric Engineering and Remote Sensing, 76, 421‐427 801