Dendroglaciological reconstruction of late-Holocene glacier activity at White and South Flat glaciers,Boundary Range,northern British Columbia Coast Mountains,Canada

Abstract

Holocene glacier activity in the northern British Columbia Coast Mountains, Canada, is described following investigations in the recently deglaciated forefields of White and South Flat glaciers spilling from the Cambria Icefield. Glacially overridden stumps and detrital wood was radiocarbon and dendrochronologically dated to describe an advance between ad 250 and 650. Expansion and thickening of White Glacier by ad 765 resulted in creation of an ice-dammed lake in South Flat valley that persisted until ad 1080. Following this the lake drained, before refilling in the early ‘Little Ice Age’ prior to ad 1300. Shortly thereafter White and South Flat glaciers became confluent and flowed over the valley side toward White Lake. The characteristics of the site and the preservation of 1500 yr old deposits indicate that the two glaciers remained confluent throughout the remainder of the ‘Little Ice Age’, only separating following terminal retreat early in the 20th century. The late-Holocene glacial history of White and South Flat glaciers appears synchronous with those of other glaciers in northern portion of Pacific Northwest.

Keywords

British Columbia Coast Mountains Cambria Icefield dendrochronology dendroglaciology

Introduction

The rapid retreat and downwasting of glaciers in the Pacific Northwest (PNW) of North America over the last few decades has exposed land surfaces and glacially killed tree remains that provide insights into periods of glacier activity during the Holocene epoch. These discoveries provide rich evidence of glacial activity and highlight a complex story of glacial advance and retreat. These findings serve to increase our understanding of glacier expansion within the Neoglacial period in this region (Clague and Matthews, 1992; Denton and Stuiver, 1966; Ryder and Thomson, 1986).

Evidence for Holocene glacier activity is widespread in the British Columbia (B.C.) Coast Mountains. Recent investigations show that glaciers in this region repeatedly advanced and retreated over the Holocene (Calkin et al., 2001; Menounos et al., 2009; Wiles et al., 2008). Eight periods of expansion are recognized: at c. 8000 ¹⁴C yr BP (Menounos et al., 2009); between 6000 and 5000 ¹⁴C yr BP (Ryder and Thomson, 1986); at 4900 ¹⁴C yr BP (Menounos et al., 2009); at c. 4200 ¹⁴C yr BP (Osborn et al., 2007); at c. 3000 ¹⁴C yr BP (Ryder and Thomson, 1986); at c. 2300 ¹⁴C yr BP (Smith and Koehler, 2011); at c. 1700 to 1500 ¹⁴C yr BP during the First Millennium Advance (FMA: Reyes et al., 2006); and, during the ‘Little Ice Age’ (LIA) from c. 900 to 100 yr BP (Larocque and Smith, 2003). Despite this growing body of evidence, there remain regions in the Coast Mountains where little or no research has been completed (Menounos et al., 2009).

This paper describes the late-Holocene history of two glaciers flowing from the relatively unstudied Cambria Icefield (Figure 1). Reconnaissance investigations in July 2008 led to the discovery of fresh exposures of subfossil wood buried in and below glacial sedimentary units. Additional sites and detrital subfossil wood deposits were discovered in July 2009. Dendroglaciological research methodologies were applied to describe the ice front behavior of these two glaciers and to consider the climatological context of their Holocene behavior.

Figure 1.

Location of White Glacier (A) and South Flat Glacier (B) in the Cambria Icefield, B.C.

Study site

White and South Flat glaciers are outlet glaciers along the eastern Cambria Icefield situated 35 km southeast of Stewart, B.C. (Lat 55°47′N, Long 129°29′W; Figure 1). Located within the Boundary Ranges, the icefield covers 715 km² and includes within its boundaries the summit of Treble Mountain at 1830 m a.s.l. A number of large outlet glaciers spill into surrounding valleys to present-day terminus positions at 885 to 510 m a.s.l. High-elevation montane forests surrounding the icefield are characterized by mixed stands of mountain hemlock (Tsuga mertensiana [Bong.] Carr.) and subalpine fir (Abies lasiocarpa [Hook.] Nutt.; Meidinger and Pojar, 1991). The upper treeline is found at around 1500 m a.s.l.

Within the last century the outlet glaciers within the Cambria Icefield area have down-wasted and receded between 2.5 and 6 km upvalley from LIA terminal positions at 815 to 380 m a.s.l. McConnell (1913) recorded 50 m of retreat at the north-facing Bromley Glacier from 1910 to 1911; Field (1975) notes that the glacier continued to retreat 30 m/yr up valley from 1911 to 1946. Matthews (1965) reports on the retreat of Bear River Glacier in the 1940s and 1950s and the impoundment of ice-proximal Strohn Lake. In the adjacent Todd Icefield area, Jackson et al. (2008) recorded terminus retreat rates at five glaciers from 1974 to 1997 at rates varying from 9 to 76 m/yr.

White Glacier

White Glacier (Lat 55°51′N, Long 129°31′W; Figure 2) flows eastward from the Cambria Icefield and in 2009 terminated at 820 m a.s.l. adjacent to South Flat River, a tributary to White River. The recently deglaciated terrain in front of White Glacier is characterized in the north forefield area by massive unconsolidated sedimentary deposits eroded by abandoned stream channels and in the southern forefield area by exposed bedrock outcrops.

Figure 2.

Site map of White and South Flat glaciers and study sites from 1946 aerial photo data.

The LIA extent of White Glacier is demarcated by a terminal moraine 2.5 km from the 2008 glacier terminus at 715 m a.s.l. The moraine dams White Lake and is colonized by Sitka alder (Alnus sinuata [Regel] Rydb.). As the glacier advanced, retreated, and periodically blocked meltwater flowing down-valley from South Flat Glacier to the north, it is likely that a glacially dammed lake formed adjacent to White Glacier and drained a number of times. Aerial photographs from 1946 (A12276) show grounded icebergs north of White Glacier. During geological investigations in the early 1960s, Grove (1973) described a glacially dammed lake close to the confluence of White and South Flat glaciers. Field investigations in 2008 found evidence suggesting the very recent draining of a glacially dammed lake

South Flat Glacier

South Flat Glacier (Lat 55°51′N, Long 129°27′W; Figure 2) is a distributary of the larger Flat Glacier (unofficial name) that flows from the Cambria Icefield to drain into South Flat River. South Flat Glacier spills over the hydrological divide south into South Flat valley. Downvalley of the 2008 glacier terminus, the valley is characterized by prominent bedrock outcrops and extensive bedded sedimentary units. Cross-sections in abandoned outwash channels show the latter unit consists of >10 m thick silt-dominated glaciolacustrine deposits, with interbedded units of steeply dipping sand and gravel delta deposits. The sedimentary units terminate down-valley adjacent to an interlobate moraine marking the confluent late-LIA extent of White and South Flat glaciers.

Research methods

Tree-ring records were collected from living trees and subfossil wood remains exposed at nine sites (Figure 2). Increment core samples were collected from a stand of living mountain hemlock trees positioned west of the 2008 terminus of South Flat Glacier. A chainsaw was used to collect cross-sectional samples from subfossil stumps and detrital wood. After air drying and mounting the samples were sanded to a fine polish at the University of Victoria Tree-Ring Laboratory. The samples were scanned and the total ring-widths measured to the nearest 0.01 mm using WinDENDRO™ software. For samples with narrow annual rings, the ring-widths were measured to the nearest 0.01 mm with MeasureJ2X (VoorTech Consulting 2009) using a Velmex-stage equipped with a microscope and CCD video display.

The living tree-ring samples were internally cross-dated using CDendro software (Ver. 7.1, Cybis Elektronic & Data AB). Following this the International Tree Ring Database (ITRDB) software program COFECHA was used to verify the cross-dating using block correlations between each tree-ring series and the master chronology (Holmes et al., 1986). COFECHA correlations were calculated using a 50-year segment length lagged successively by 25 years at a one-tailed 99% confidence level (Grissino-Mayer, 2001). A living tree-ring chronology was constructed using double detrending options (negative exponential option 2 and cubic smoothing spline option 4) in the ITRDB software program ARSTAN (Cook and Holmes, 1986) to retain low-frequency variability within the resulting chronology (Cook et al., 1990). To create a single dimensionless value for each growth year, each ring-width index series was prewhitened using autoregressive and moving average models to remove autocorrelation effects (Biondi and Swetnam, 1987; Cook, 1985).

The species of subfossil wood samples was identified using a 40× microscope and a standard reference key (Hoadley, 1990). Site-specific floating tree-ring chronologies were constructed from the subfossil samples following established cross-dating protocols (Stokes and Smiley, 1968) and CDendro software. An attempt was first made to determine the absolute age of each floating chronology by cross-dating to the standardized living chronology. Where this failed, calibrated ages determined from perimeter wood samples submitted to Beta Analytic Inc. for standard radiocarbon dating were assigned to the floating chronologies (Luckman, 1995; Smith and Lewis, 2007). A calendar age is represented by taking the midpoint of the calibrated age range dates, except in the case of multiple age intercepts where all possible ages are reported (Koehler and Smith, 2011).

Observations

Tree-ring chronology

Twenty-seven ring-width series from 17 trees were used to construct a living tree-ring chronology spanning 373 years (ad 1635–2007). The duration and cross-dating potential of the chronology was enhanced by including it with a regional chronology constructed from samples previously collected at nearby Surprise Glacier (ad 1596–2007; Jackson et al., 2008) and Windy Pass (ad 1613–2004; Figure 1). The three records strongly cross-date (series correlation 0.535–0.595) to form a chronology (n= 87 series) spanning 412 years (ad 1596–2007) with an expressed population signal (EPS) with a 0.8 cutoff at ad 1680.

Dendroglaciological samples

Glacially sheared stumps and woody debris were found at six locations proximal to White Glacier and three sites downstream of South Flat Glacier (Figure 2).

Site 1 was located c. 750 m from the 2007 terminus of White Glacier within a 25 m high abandoned channel cutbank at 653 m a.s.l. (Figure 2). Distinguished by three stratigraphic units, the lower unit (Unit 1A; Figure 3) consists of a massive 10 m thick matrix-supported till with an upper surface truncated by a laterally extensive wood mat. The mat extends for c. 15 m and contains glacially transported bole, branch and stump fragments inter-mixed with till. The wood mat is overlain by a weakly bedded 7 m thick till unit (Unit 1B; Figure 3) capped by an oxidized sedimentary horizon dipping up-slope 20° west–east. At this juncture a thin composite horizon consisting of small detrital wood pieces and gravels extends across the cut face. Resting on the surface of this horizon is a 7 m thick till unit (Unit 1C; Figure 3) whose upper surface features small-scale northwest-trending flutes. Cross-sections from boles found at the boundary between Units 1A and 1B indicate the trees were from 126 to 160 years old when killed. An age of ad 605 was assigned to the outermost perimeter rings of sample WG01_01 (Table 1). Cross-dating of this sample to WG01_02 and WG01_03 (Table 1) led to the construction of an 221 year long floating mountain hemlock chronology extending from c. ad 247 to 468 (Chronology 1, Table 2).

Figure 3.

Looking north towards site 1. Three units are highlighted: A 10 m thick matrix-supported till (1A), wood samples collected from the wood mat (WG01_01) on the surface of 1A date to ad 605. This unit is overlain by a 7 m thick till unit (1B) capped by an orange-stained sedimentary horizon. Resting on the surface of this horizon is a 7 m thick till unit (1C) whose upper surface features small northwest-trending glacial flutes.

Table 1.

Summary of conventional age radiocarbon dated samples from White and South Flat glaciers. A calander age is represented by taking the midpoint of the calibrate age range dates (from Koehler and Smith, 2011).

Sample ID	BETA ID	Species^a	Perimeter rings	¹⁴C age yr BP	2σ calibration age range (ad)	2σ calibration date BP	Assigned age (years ad)
WG01_01	262876	mh	98	1460 ± 40	550–660	1400–1290	605
WG02_03	262877	mh	104	1550 ± 40	420–600	1530–1350	510
WG03_10	273508	saf	34	1360 ± 40	600–680	1350–1270	640
WG03_14	262878	mh	68	1530 ± 40	420–610	1520–1340	515
WG05_02	262879	saf	76	1360 ± 50	570–680	1380–1270	625
WG06_04	250505	saf	62	1400 ± 40	590–670	1360–1280	630
WG07_01	250501	mh	120	1280 ± 50	650–880	1300–1070	765
WG08_12	273508	mh	67	960 ± 40	1000–1160	950–790	1080
WG09_16	250503	mh	121	810 ± 40	1160–1280	790–670	1220
WG09_18	250504	saf	46	610 ± 60	1280–1430	670–520	1355

saf: subalpine fir; mh: mountain hemlock. Species were identified by their gross anatomical characteristics (Hoadley, 1990).

All radiocarbon dates and 2σ ranges calculated by Beta Analytic Inc. using INCAL04.

Table 2.

The floating tree-ring chronologies created and their conventional age dates assigned after unsuccessful cross-dating into the living tree-ring chronology. If the sample was used in the FMA Chronology then the perimeter age date is recorded.

Sample no.	¹⁴C age yr (±2σ BP)	Number of rings	Correlation with master	Perimeter age (FMA Chronology)
Floating Chronology Site 1
WG01_01A	1460 ± 40	122	0.292
WG01_02A		160	0.232	468
WG01_02B		149	0.468	461
WG01_03B		91	0.354	417
Floating Chronology Site 2
WG02_01A		179	0.321	488
WG02_01B		185	0.375
WG02_02A		237	0.490	555
WG02_02B		185	0.625	511
WG02_03A	1550 ± 40	193	0.375	517
Floating Chronology Site 3
1 WG03_10A	1360 ± 40	316	0.481	548
WG03_10B	1360 ± 40	314	0.409	548
WG03_11A		149	0.331
WG03_11B		112	0.489
2 WG03_02A		217	0.621	582
WG03_02B		217	0.545	582
WG03_03A		153	0.224
WG03_06A		110	0.211
WG03_06B		119	0.370
3 WG03_14A	1530 ± 40	227	0.539	592
WG03_14B	1530 ± 40	199	0.539	560
Floating Chronology Site 4
WG04_01A		110	0.546
WG04_01B		94	0.790
WG04_02A		89	0.610	634
WG04_02B		106	0.438	634
WG04_03A		167	0.353
WG04_03B		167	0.313
WG04_04A		93	0.521
WG04_04B		99	0.595
Floating Chronology Site 5
WG05_01A		208	0.447	554
WG05_01B		205	0.422	552
WG05_02B		100	0.314
Floating Chronology Site 6
WG06_04A		63	0.786	559
WG06_04B		63	0.619
WG06_08A		63	0.560	601
WG06_08B		72	0.498	612
WG06_09A		59	0.334	555
WG06_09B		72	0.330
Floating Chronology Site 7
1 SF07_01A	1280 ± 40	456	0.425
SF07_01B	1280 ± 40	454	0.425
2 SF07_02A		208	0.560
SF07_02B		206	0.644
SF07_04A		166	0.402
SF07_04B		181	0.250
SF07_05A		101	0.403
SF07_05B		105	0.594
Floating Chronology Site 8
SF08_09A		114	0.209
SF08_10A		93	0.438
SF08_10A		95	0.566
SF08_11A		115	0.435	631
SF08_11B		119	0.408
SF08_12A		60	0.570
SF08_12B		77	0.596
Floating Chronology Site 9-lower
SF09_16A	810 ± 40	219	0.468
SF09_16B	810 ± 40	180	0.468
Floating Chronology Site 9-upper
SF10_17B		113	0.336
SF09_18A	610 ± 60	119	0.643
SF09_18B	610 ± 60	133	0.624

Site 2 is located 25 m upstream from Site 1 at 673 m a.s.l. (Figure 2). Exposed in a vertical cutbank is a 20 m thick section of till containing 2–3 m long boles with masticated surfaces. Cross-sections were collected from four boles positioned 12 and 17 m above the abandoned channel floor. The samples contain from 185 to 237 annual rings, and three cross-date to form a 250 year long floating mountain hemlock chronology (Chronology 2, Table 2). Radiocarbon dating of perimeter rings from WG02_03 indicate the chronology extends from c. ad 305 to 555 (Table 1).

Site 3 is positioned below the proximal face of a prominent 10 m high glacially eroded bedrock outcrop at 679 m a.s.l. (Figure 2). The colluvial footslope consists of diamicton intermixed with partially buried broken and fragmented bole segments up to 6 m in length. Cross-sections from 15 samples led to the construction of two chronologies. A subalpine fir chronology (Chronology 3-1, Table 2) extending over a 316 yr interval is anchored by perimeter wood dating to ad 640 (WG03_10; Tables 1 and 2). A mountain hemlock chronology from the same location includes three series with between 119 and 217 annual rings (Chronology 3-2; Table 2), and extends over a 217 yr period. While a radiocarbon date was not assigned to the samples included within the latter chronology, it is assumed that the trees were killed at the same time (ad 232 to 548) as those incorporated into Chronology 3-1 given that the remains were intermixed and buried (Table 2).

Site 4 is located 25 m west of Site 3 at 675 m a.s.l. on the east face of a prominent bedrock outcrop. The site was ice-free prior to 1946 and contains 0.5–1 m long bole segments (99 to 110 rings) pressed into a small pocket of organics. Three subalpine fir samples cross-date to form a floating 110 yr long chronology (Chronology 4, Table 2) that cross-dates to Chronology 3-3 at Site 3 (Table 2), indicating that they were killed c. ad 515 (Table 1).

Site 5 is located c. 20 m north of sites 3 and 4, in an 6 m deep east–west facing bedrock gully at 660 m a.s.l. A large accumulation of broken and fragmented detrital wood appears to have been pushed into the gully as White Glacier advanced over the surrounding bedrock surface. Although most of the wood at Site 5 was rotted and in poor condition, cross-sections from two boles (WG05_01 and WG05_02; Table 2) cross-date to form a 205 yr long chronology killed in c. ad 625 (Tables 1 and 2).

Site 6 is found north of White Glacier at 1002 m a.s.l. on the west side of a prominent bedrock ridge separating the White and South Flat glacier forefields, where a pocket of till was lodged into a bedrock gully. Cross-sections retrieved from five partially buried 2–7 m long boles (63–72 rings) cross-date to form an 81 year long floating chronology killed c. ad 640 (WG06_04; Tables 1 and 2).

Site 7 is located 2 km downstream from the 2007 terminus of South Flat Glacier at 762 m a.s.l. where meltwater incised through over 70 m of till and glaciolacustrine sediment. A sheared stump with 456 annual rings (WG07_01; Table 1) in growth position exposed along the channel bank of South Flat River was killed c. ad 765 (Table 1). Cross-sections from three nearby detrital logs cross-date to form Chronology 7-2 (Table 2), but do not cross-date to WG07-01.

Site 8 is found 900 m from the 2007 terminus of South Flat Glacier at 820 m a.s.l. Adjacent to a prominent bedrock gorge is a 25 m thick exposure of steeply dipping laminated sands and gravels capped by till. The latter extends upwards to a non-conformable contact with fine-grained horizontally bedded rhythmites. Numerous woody fragments and bole segments were exposed within the upper section of the lower laminated unit. Cross-sections collected from five small logs cross-date (59-117 rings; Table 2) and sample WG08_11 indicates all the trees died c. ad 1080 (Table 1).

Site 9 refers to a 60 m high vertical cross-section located c. 50 m above Site 7 consisting of a horizontally bedded laminates with alternating units of gravel and silty-sands. Woody horizons at several levels appear matched by equivalent units visible on the other the side of the valley. Cross-sections from detrital wood collected at 840 m a.s.l. (130–219 rings; Chronology 9 lower, Table 2) indicates the trees were killed in ad 1200 (WG09_16; Table 1). Samples collected from two detrital boles at 860 m a.s.l. (113–133 rings; Chronology 9 upper, Table 2) indicates these higher elevation trees were killed and buried in c. ad 1355 (WG09_18; Tables 1 and 2).

Results

Twelve site- and species-specific floating chronologies were created from subfossil wood found at recently exposed sites in the White and South Flat glacier forefields (Table 2). Cross-dating to the regional mountain hemlock and an archived subalpine fir chronology (Jackson et al., 2008) was unsuccessful, indicating that the subfossil samples were killed prior to ad 1596 and ad 1649, respectively.

Seven floating chronologies cross-date to one another and were combined to form the ‘Cambria’ chronology (Table 2; Figure 4). The chronology spans a 393 year interval and includes numerous radiocarbon-dated samples (Tables 1 and 2) indicating the trees were killed at various times between ad 251 to 664 (Figure 4).

Figure 4.

First Millennial Advance chronology anchored by conventional radiocarbon date from WG03_10 (Table 1). The chronology consists of samples from White and South Flat glaciers, as well as previously collected samples from the Todd Icefield area (from Jackson et al., 2008). The chronology extends from ad 251 to 664 (393 years) and consists of 38 tree-ring records.

The floating ‘Cambria’ chronology was subsequently cross-dated with 15 floating subfossil samples previously collected in the nearby Todd Icefield area (Jackson et al., 2008). The resulting ‘Stewart’ chronology contains 38 tree-ring records, spans a 393 year period, and includes six radiocarbon-dated perimeter wood samples (Table 1). Perimeter wood from WG03_10 records the most recently killed sample (ad 600–680; Table 1) and was used anchor to the chronology (Figure 4). This adjustment resulted in the kill-dates of the five remaining radiocarbon samples falling within 2σ of the midpoint of their calibrated age range (Table 1).

Interpretation

Field investigations were used to document an FMA advance of White Glacier from c. ad 250 to 650; as well as the early-LIA interactions of both White and South Flat glaciers. No evidence was found to describe glacier activity earlier in the Holocene, despite nearby records detailing extensive advances c. 3000 years ago (Clague and Matthews, 1992; Jackson et al., 2008).

Evidence from the White Glacier forefield (sites 1–6) describes an FMA ice front advance from ad 486 to 664 into a standing forest composed of mountain hemlock and subalpine fir trees. Shortly before ad 486 the glacier crossed South Flat River and began moving up the west-facing valley side overwhelming trees. The glacier reached Site 1 in ad 486 and deposited wood at Site 2 in ad 596. The minimum rate of advance over this interval was 4.4 m/yr.

Following this the snout of White Glacier continued to thicken and extend up the valley site, reaching Site 3 by c. ad 610. At about the same, White Glacier was pushing woody detritus into the gully at Site 5. The glacier snout reached Site 4 by ad 664, having advanced up the bedrock slope from the vicinity of Site 3 at a minimum of 0.4 m/yr.

The distal position of the woody detritus at Site 4 indicates that White Glacier advanced over the ridge crest and was flowing downslope toward White Lake after ad 664. While no dendroglaciological evidence was recovered describing the maximum extent of this advance, cross-dating established that White Glacier was killing trees at both sites 5 (660 m a.s.l.) and 6 (915 m a.s.l.) c. ad 625.

FMA expansion and thickening of White Glacier eventually resulted in an ice dam that blocked the drainage of meltwater flowing from South Flat Glacier down South Flat valley (Figure 5). The timing and nature this ice dam is constrained by the in-situ stump located at Site 7 (WG07_01; Table 1). Killed when submerged by proglacial lake water in ad 765 (Figure 5), the tree was eventually engulfed by laminated glaciolacustrine sediments and then overtopped by prograding delta sediments as South Flat Glacier advanced down-valley.

Figure 5.

Schematic of the expansion of White Glacier and the subsequent interaction of White Glacier with South Flat Glacier between c. ad 600 and the early ‘Little Ice Age’. In panel 1, the glacier is advancing to Site 3 with South Flat River flowing beneath the glacier. In panel 2, the glacier has thickened sufficiently to force the ponding of a lake, killing sample 07_01 at site 7. South Flat River subsequently flows around White Glacier. Panel 3 depicts the advance of South Flat Glacier, causing the formation of a prograding delta, deposition of sediments and death of trees at site 9. White and South Flat glacier are not confluent until the late ‘Little Ice Age’ (panel 4), where all of the South Flat valley and White Glacier forefield is covered with glacial ice.

Abandoned bedrock channels at c. 800 m a.s.l. indicate that water from the lake spilled over the intervening ridge to White Lake at a late high-elevation stage (Figure 5). The existence of an ice-dammed lake in ad 765, persisting long enough for thick sedimentary units to accumulate, indicates the glacier remained in this advanced position until sometime after ad 800. Wood fragments contained within the lower deltaic unit at Site 8 indicate the lake may have persisted until ad 1080; although it seems more likely this deposit is associated with ice blockage resulting from the LIA advance of White Glacier.

The initiation of LIA glacier activity at White and South Flat glaciers is recorded by detrital wood found within the thick unit of glaciolacustrine deposits that unconformably overlie FMA-aged sediments at Site 9. The large bole segments recovered within alternating sand and gravel units at the base of the unit (840 m a.s.l.) are interpreted to record a proglacial subaerial setting (WG08_12; Table 1). Overlying this deposit is a thick unit of coarsely laminated sand facies that records sedimentation in a shallow ice- or moraine-dammed lake associated with the extension of White Glacier into South Flat valley. Broken bole fragments dating to ad 1355 at 860 m a.s.l. are oriented down-valley and record flows originating from South Flat Glacier. Overlying this unit are coarsening-upward glaciofluvial units directly associated with the late-LIA moraines. The glaciers separated prior to 1949 (Figures 2 and 5). White Glacier continued to straddle South Flat River until after 2001 and there is evidence for the filling and drainage of a succession of short-lived ice-dammed lakes.

Discussion and synthesis

Although previous research in the region indicates the FMA is distinguished by an early (ad 375) and late (ad 605) phase (Jackson et al., 2008), these investigations provide evidence for only a single continuous advance of White Glacier. The dendroglaciological findings from the site describe the advance of White Glacier in South Flat valley before ad 486. Over the following 250 years or so a gradual thickening of the glacier eventually blockaded the valley and led to the creation of an ice-contact lake before ad 765.

Thick bottomset and delta foreset sedimentary units describe an interval of sustained infilling of the lake that persisted until after ad 800. White Glacier appears to have thinned sufficiently, before ad 1080, to allow the lake to partially drain around or beneath the glacier snout. Following this White Glacier readvanced and by ad 1080 blocked South Flat valley to create a second ice-contact lake that persisted until after ad 1355. Overlying till deposits indicate that White and South Flat glaciers were confluent in South Flat valley following this; a state that may have persisted until the end of the LIA in the last century.

The reconstructed FMA and LIA glacial history of White and South Flat glaciers is comparable with that reported at other sites in the region (Reyes et al., 2006). Clague and Matthews (1992) note that Frank Mackie Glacier was in an advanced state from ad 320 to 570. In the nearby Todd Icefield area, Jackson et al. (2008) report that several glacier fronts were advancing and expanding down valley from ad 230 to 650. In the Andrei Icefield area, 150 km northwest of the study site, Forrest Kerr, Scud and South More glaciers were actively advancing into standing forests at c. ad 460 (Craig et al., 2011; Lewis and Smith, 2005). Further afield in Alaska, coeval FMA kill dates have been found at Beare (Johnson et al., 1997) and Icy Bay glaciers (Barclay et al., 2006) from ad 420 to 670, and at the Llewellyn Glacier in the Juneau Icefield from ad 1030 to 1150 (Koch and Clague, 2011).

The early-LIA activity of South Flat and White glaciers appears contemporaneous with activity documented at nearby Beare River (ad 960; Spooner et al., 2005) and Surprise glaciers (ad 1390; Jackson et al., 2008). Corresponding glacier advances have been documented in the Todd Icefield (Jackson et al., 2008) at ad 1450 and in the Stikine area at Scud Glacier at c. ad 1380 (Ryder, 1987). At Tebenkof Glacier in southern Alaska, a comparable early LIA advance was underway at about ad 1320 (Barclay et al., 2009).

The apparent regional synchronicity of these FMA and early-LIA advances prompted an examination of ring-width variability within the two climate-sensitive tree-ring chronologies constructed as part of this study. To highlight temporal rhythms retained within the subfossil and living chronologies, an interactive wavelet analysis (http://paos.colorado.edu/research/wavelets; Torrence and Compo, 1998) with a Gaussian 2 function and a 10% white noise reduction was completed.

As shown in Figure 6, both chronologies retain a repetitive 8 year event thought to be associated with climate forcing mechanisms described by the El Niño Southern Oscillation (ENSO). This mode of variability persists throughout the duration of the live chronology (Figure 6a), but appears to weaken in the Stewart chronology as FMA glacier expansion began prior to ad 460 (Figure 6b). A 32 year event thought to be associated with the Pacific Decadal Oscillation (PDO) is prominent within the live chronology; this pattern is more weakly expressed within the FMA chronology after ad 425 until ad 550. These periodicities are synchronous to those reported further south in the Mt Waddington region (Larocque and Smith, 2004) and the Pacific Northwest (Gedalof and Smith, 2001). The live chronology was also compared with a reconstructed PDO index (Gedalof and Smith, 2001). The two series have a 387 year overlap and show an overall positive relationship between PDO index and tree-ring growth. Across this region cooler than normal JJA air temperatures characterize the intervals between 1698 and 1706, 1876 and 1886, and 1970 and 1976. Warmer than normal JJA air temperatures occurred from 1715 to 1726, 1800 to 1820, 1900 to 1920 and 1945 to 1955. These alternating warm–cold temperature regimes are in step with phases changes in the PDO and highlight the effect of both ‘warm’ and ‘cold’ PDO phases changes on JJA air temperature in this region.

Figure 6.

The wavelet power spectrum for the (a) Live Chronology, and (b) the Stewart Chronology. The contour levels are chosen so that 75%, 50%, 25%, and 5% of the wavelet power is above each level, respectively. The cross-hatched region is the cone of influence, where zero padding has reduced the variance. The black contours demark the 15% significance level, using a white-noise background spectrum (Torrence and Compo, 1998).

Within shorter timescales, the year-to-year radial growth of trees in this region positively correlated to mean summer (June–July–August) air temperature. These temporal signals retained within the living and subfossil chronologies indicate that climate forcing events originating over the Pacific Ocean have persistently influenced the radial growth of trees in this region. As glacier mass balance changes in this region are principally influenced by climate variables (Larocque and Smith, 2005), the correspondence of shifts in the character of these relationships prior to the FMA and early-LIA advances documented in this study hint at the role that La Niña events may have played in generating positive glacier mass balance states and glacier advances over the last two millennium (c.f. Villalba et al., 2001; Watson et al., 2006). Documentation of enhanced PDO activity after ad 550 and in the late-LIA period (ad 1800–1900), likely to be associated with higher winter precipitation and cooler summers, may similarly reflect significant intervals of positive mass balance (Bitz and Battisti, 1999; Lewis and Smith, 2004). Observation of the persistent influence of phase pairings of La Niña/negative PDO years on glacier mass balance over the last two millennium confirms the findings of shorter duration studies elsewhere in the western Canadian Cordillera (Watson et al., 2006).

Proxy evidence of this relationship comes from western hemlock (Tsuga heterophylla [Raf.] Sarg.) pollen deposited in Pyramid Lake in the Cassiar Mountains of B.C. (Spooner et al., 2003). Increases in the influx of coastal pollen types beginning c. 1500 yr BP is interpreted to be a consequence of changes in regional air-mass circulation patterns that brought cooler and wetter weather to PNW (Overland et al., 1999). If this interpretation is correct, the FMA advances recorded throughout PNW at this time may be associated with persistence of deeper Aleutian Low pressure systems (Trenberth, 1990).

Summary

Glaciers flowing from the Cambria Icefield in northwestern B.C. have receded and down-wasted significantly in the last 60 years. Investigations completed in the recently deglaciated forefields of White and South Flat glaciers led to the discovery of two late-Holocene intervals of glacier expansion. Dendroglaciological evidence from the White Glacier forefield describes an FMA ice-front advance from ad 486 to 664. Expansion and thickening of White Glacier by ad 765 resulted in creation of an ice-dammed lake in South Flat valley that may have persisted until ad 1080, after which the lake drained before refilling in the early LIA. After ad 1300 White and South Flat glaciers were confluent and flowed over the valley side toward White Lake. The characteristics of the site and the preservation of FMA-aged deposits indicate that the two glaciers remained confluent throughout the remainder of the LIA, only separating following terminal retreat early in the 20th century.

The late-Holocene glacial history of White and South Flat glaciers appears synchronous with those of the other glaciers in northern portion of PNW. An assessment of living and subfossil tree-ring records from the Cambria Icefield area was used to infer the climate forcing mechanisms that resulted in the documented FMA and early-LIA advances. Persistent temporal relationships to the ENSO and PDO climate indices are reflective of the strong maritime influence that the Pacific Ocean has on glacier mass balance regimes in the region. Whereas decadal–interdecadal shifts in these relationships result in ice front fluctuations, it seems likely that sustained events like the FMA arose following pronounced reconfigurations of regional air-mass circulation patterns.

Footnotes

Acknowledgements

We extend our thanks to the UVTRL for field assistance in 2008 and 2009. We thank two anonymous reviewers for their helpful comments and suggestions to earlier drafts of our paper.

Funding

D Smith received project support through a grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) and from an award made by the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS) to the Western Canadian Cryospheric Network.

References

Barclay

Wiles

Calkin

(2009) Holocene glacier fluctuations in Alaska. Quaternary Science Reviews 28: 2034–2048.

Barclay

Calkin

. (2006) A revised and extended Holocene glacial history of Icy Bay, southern Alaska, U.S.A. Arctic, Antarctic and Alpine Research 38: 153–162.

Biondi

Swetnam

(1987) Box-Jenkins models of forest interior tree-ring chronologies. Tree-Ring Bulletin 4: 771–795.

Bitz

Battisti

(1999) Interannual to decadal variability in climate and the glacier mass balance in Washington western Canada and Alaska. Journal of Climate 12: 3181–3196.

Calkin

Wiles

Barclay

(2001) Holocene coastal glaciation of Alaska. Quaternary Science Reviews 20: 449–461.

Clague

Matthews

(1992) The sedimentary record and Neoglacial history of Tide Lake, northwestern British Columbia. Canadian Journal of Earth Sciences 29: 2383–2396.

Cook

(1985) A time-series analysis approach to tree-ring standardization. PhD thesis, University of Arizona.

Cook

Holmes

(1986) Guide for computer program ARSTAN (Adapted from Users Manual for Program ARSTAN). In: Holmes

Fritts

(eds) Tree-Ring Chronologies of Western North America: California Eastern Oregon and Northern Great Basin. Tucson AZ: University of Arizona Laboratory of Tree Ring Research.

Cook

Shiyatov

Mazepa

(1990) Estimating the mean chronology. In: Cook

Kairiukstis

(eds) Methods of Dendrochronology: Applications in the Environmental Sciences. Boston MA: Kluwer Academic Publishers, pp. 123–132.

10.

Craig

Smith

Lewis

(2011) Dendroglaciological investigations at South More Glacier, Northern British Columbia Coast Mountains. 2011 Annual Meeting of the Western Division of the Canadian Association of Geographers. 10–12 March 2011. Simon Fraser University, Burnaby BC. Abstract.

11.

Denton

Stuiver

(1966) Neoglacial chronology North Eastern St Elias Mountains Canada. American Journal of Science 264: 577–599.

12.

Field

(1975) Chapter 1. Glaciers of the Coast Mountains: Boundary Ranges (Alaska, British Columbia, and Yukon Territory). In: Field

(ed.) Mountain Glaciers of the Northern Hemisphere. Volume 2. Hanover NH: Cold Regions Research and Engineering Laboratory, pp. 11–141.

13.

Gedalof

Smith

(2001) Dendroclimatic response of mountain hemlock (Tsuga mertensiana) in Pacific North America. Canadian Journal of Forest Research 31(2): 322–332.

14.

Grissino-Mayer

(2001) Evaluating cross-dating accuracy: A manual and tutorial for the computer program COFECHA. Tree-Ring Research 57: 205–221.

15.

Grove

(1973) Detailed geological studies in the Stewart complex, northwestern British Columbia. PhD thesis. McGill University.

16.

Hoadley

(1990) Identifying Wood: Accurate Results with Simple Tools. Newtown CT: Taunton Press.

17.

Holmes

Adams

Fritts

(1986) Tree-ring Chronologies of Western North America: California. Eastern Oregon and Northern Great Basin. Laboratory of Tree-Ring Research Chronology Series VI. Tucson AZ: University of Arizona Press.

18.

Jackson

Laxton

Smith

(2008) Dendroglaciological evidence for Holocene glacial advances in the Todd Icefield area, Northern British Columbia Coast Mountains. Canadian Journal of Earth Sciences 45: 83–98.

19.

Johnson

Wiles

Frank

(1997) Three millennia of glacial fluctuations from Beare Glacier, eastern Gulf of Alaska. Geological Society of America Abstracts with Programs 29: 22–23.

20.

Koch

Clague

(2011) Extensive glaciers in northwest North America during Medieval time. Climate Change 107: 593–613.

21.

Koehler

Smith

(2011) Late-Holocene glacial activity in Manatee Valley, southern Coast Mountains, British Columbia, Canada. Canadian Journal of Earth Sciences 48: 603–618.

22.

Larocque

Smith

(2003) Little Ice Age glacial activity in the Mt Waddington area British Columbia Coast Mountains Canada. Canadian Journal of Earth Sciences 40: 1413–1436.

23.

Larocque

Smith

(2004) A dendroclimatological reconstruction of climate since AD 1700 in the Mt. Waddington area, British Columbia Coast Mountains, Canada. Dendrochronologia 22: 93–106.

24.

Larocque

Smith

(2005) ‘Little Ice Age’ proxy glacier mass balance records reconstructed from tree rings in the Mt Waddington area, British Columbia Coast Mountains, Canada. The Holocene 15: 748–757.

25.

Lewis

Smith

(2004) Dendroglaciological mass balance reconstruction Strathcona Provincial Park, Vancouver Island, British Columbia, Canada. Arctic Antarctic and Alpine Research 36: 598–606.

26.

Lewis

Smith

(2005) Dendroglaciological evidence of late Holocene glacier activity at Forrest Kerr Glacier, Andrei Icefield, northern British Columbia Coast Mountains. Annual Meeting of the Canadian Association of Geographers, 31 May–4 June 2005, University of Western Ontario, London, Ontario.

27.

Luckman

(1995) Calendar-dated early ‘Little Ice Age’ glacier advance at Robson Glacier, British Columbia, Canada. The Holocene 51: 49–59.

28.

McConnell

(1913) Portions of Portland Canal and Skeena Mining Divisions, Skeena District, BC. Ottawa: Geological Survey of Canada, Memoir 32.

29.

Matthews

(1965) Two self-dumping ice-dammed lakes in British Columbia. Geographical Review 55: 46–52.

30.

Meidinger

Pojar

(1991) Ecosystems of British Columbia. Victoria, British Columbia: British Columbia Ministry of Forests, Special Report No. 6.

31.

Menounos

Osborn

Clague

. (2009) Latest Pleistocene and Holocene glacier fluctuations in western Canada. Quaternary Science Reviews 28: 2049–2074.

32.

Osborn

Menounos

Clague

. (2007) Multi-proxy record of Holocene glacial history of the Spearhead and Fitzsimmons ranges southern Coast Mountains. Quaternary Science Reviews 26: 479–493.

33.

Overland

Adams

Bond

(1999) Decadal variability of the Aleutian Low and its relation to high-latitude circulation. Journal of Climate 12: 1542–1548.

34.

Reyes

Wiles

Smith

. (2006) Expansion of alpine glaciers in Pacific North America in the first millennium AD. Geology 34: 57–60.

35.

Ryder

(1987) Neoglacial history of the Stikine-Iskut area, northern Coast Mountains, British Columbia. Canadian Journal of Earth Sciences 24: 1294–1301.

36.

Ryder

Thomson

(1986) Neoglaciation in the southern Coast Mountains of British Columbia chronology prior to the late Neoglacial maximum. Canadian Journal of Earth Sciences 23: 273–287.

37.

Smith

Lewis

(2007) Dendroglaciology. In: Elias

(ed.) Encyclopedia of Quaternary Science. Amsterdam: Elsevier Scientific, pp. 986–994.

38.

Spooner

Haspel

Osborn

(2005) Holocene history of Bear River Glacier northern Coast Ranges British Columbia. In: Water Ice Land and Life: The Quaternary Interface. Bi-Annual Meeting of the Canadian Quaternary Association, University of Manitoba, Winnipeg, Manitoba. Abstract: A87.

39.

Spooner

Barnes

Baltzer

. (2003) The impact of air mass circulation dynamics on Late Holocene paleoclimate in northwestern North America. Quaternary International 108: 77–83.

40.

Stokes

Smiley

(1968) Tree-Ring Dating. Chicago IL: The University of Chicago Press.

41.

Torrence

Compo

(1998) A practical guide to wavelet analysis. Bulletin of the American Meteorological Society 79: 61–78.

42.

Trenberth

(1990) Recent observed interdecadal climate changes in the northern hemisphere. Bulletin of the American Meteorological Society 71: 988–993.

43.

Villalba

D’Arrigo

Cook

. (2001) Decadal-scale climatic variability along the extra-tropical western coast of the Americas over past centuries inferred from tree-ring records. In: Markgraf

(ed.) Interhemispheric Climate Linkages. Cambridge: Cambridge University Press, pp. 155–172.

44.

Watson

Luckman

(2006) Long-term relationships between reconstructed seasonal mass balance at Peyto Glacier, Canada, and Pacific sea surface temperature. The Holocene 16: 783–790.

45.

Wiles

Barclay

Calkin

. (2008) Century to millennial-scale temperature variations for the last two thousand years indicated from glacial geologic records of southern Alaska. Global and Planetary Change 60: 115–125.