Palaeoecology of postglacial treeline shifts in the northern Cascade Mountains, Canada

Palaeogeography, Palaeocimatology, Palaeoecology 141 (1998) 123—138

Palaeoecology of postglacial treeline shiftsin the northern Cascade Mountains, Canada

PALAEO

• Marlow G. Pellatt a,*, Michael J. Smith a, Roif W. Mathewes a, Ian R. Walker a,b

a Department ofBiological Sciences and Institute for Quatemaiy Research, Simon Fraser University. 8888 University Drive,

Burnaby, B.C. liSA 186, Canadab Department ofBiology, Okanagan University College, 3333 College Way, Kelowna, B.C. VI VI IC, Canada

Abstract

Received 22 July 1997; accepted 12 January 1998

Postglacial changes in vegetation and chironomid communities at a subalpine lake in the Engelmann Spruce—SubalpineFir zone in the northern Cascade Mountains, British Columbia, indicate patterns of treeline and climate fluctuationduring the Holocene. In late-glacial sediments of Cabin Lake, pollen assemblages representative of alpine vegetation andcold-stenothermous chironomids indicate cold conditions prior to the Holocene. Jn the early Holocene (10,090 to 7000 t4Cyr BP) co-occurrence of spruce—fir parkland and a warm-adapted chironomid community indicates a warm and probablydry climate. In the mid-Holocene, inferred forest closure suggests that precipitation increased, and a mixture of warm- andcold-adapted chironomids indicates temperatures warmer than present, but cooler than in the early Holocene. This periodbetween 7000 and 3200 ‘4C yr BP represents a transitional climate in which temperature gradually declined, culminating incool neoglacial conditions. This transitional interval may correspond with the ‘mesothermic period’ proposed for lowlandsites in southern British Columbia. Palaeobotanical evidence suggests that moist subalpine forest began to establisharound 4800 ‘4C yr BP with minimum temperatures and maximum precipitation between 2435 and ca. 1700 ‘4C yrBP, corresponding with neoglacial advances throughout the northern Cordillera. A cool late Holocene (3200 14C yr BPto present) is also supported by a further decline in warm-adapted chironomids. Comparisons with other study sites inthe Pacific Northwest reveal that regional climatic changes were a major factor in driving biotic changes in this area.© 1998 Elsevier Science B.V. All rights reserved.

Keywords: palaeodlimate; pollen analysis; chironomids; treeline; Cascade Mountains; British Columbia; vegetation history

1. Introduction

The postglacial vegetation and climate historyof the southern interior of British Columbia hasbeen examined at only a few locations (Alley, 1976;Mathewes and King, 1989; Hebda, 1995), all atrelatively low elevations. Palaeoecological studiesin north-central Washington (Mack et at, l978a,b,

* Fax: +1(604) 291-3496; E-mail: [email protected]

1979; Mehringer, 1985), reflect a similar lack of highelevation data. We present here a study of treelineand vegetation shifts in the subalpine EngelrnannSpruce—Subalpine Fir zone (ESSF), as an aid in theunderstanding of Holocene vegetation dynamics andpalaeoclimatic history in the southern interior. Ourstudy also allows comparisons with better-studiedcoastal localities.

Palaeoecological studies of subalpine treelineshifts on the Queen Charlotte Islands revealed three

ELSEVIER

0031-01821981$19.00 © 1998 Elsevier Science By. All rights reserved.Pit 50031-0182(98)00014-5

124 M.G. Pellatt et at. /Palaeogeography, Palaeoclimatology, Palaeoecotogy 141 (1998) 123—138

phases of vegetation and inferred climate changeover the last 10,000 radiocarbon years (Pellatt andMathewes, 1994, 1997; Pellatt, 1996). The earlyHolocene (10,000 to ca. 6000 ‘4C yr BP) on theQueen Charlotte Islands was warmer and drier thanpresent, sustaining tree species such as western hemlock (Tsuga heterophylla) and Sitka spruce (Piceasitchensis) that are now characteristic of lowlandtemperate forests. This warm/dry period is recordedat lowland study sites throughout British Columbia,southeast Alaska and northern Washington (Warner,1984; Mathewes, 1985; Mehringer, 1985; Barnoskyet at, 1987; Quickfafl, 1987; Fedje, 1993; Hebda,1995; Mann and Hamilton, 1995; Hansen and Engstrom, 1996). Subalpine conditions became established on the Queen Charlotte Islands after 6000

yr BP as the Pacific Northwest coast becamecooler and wetter. Modem vegetation and climatebecame established on the Queen Charlotte Islands,coastal British Columbia, northern Washington, andsoutheast Alaska by ca. 3500 ~4C yr BP (Fig. 1).

While coastal localities were experiencing a cooling trend in the middle Holocene, it appears thatwarm temperatures were prolonged in the interiorof British Columbia. Hebda (1995) suggests thatmid-Holocene (7000 to 4500 ‘4C yr BP) temperatures were similar to the early Holocene, with modem levels of precipitation (Fig. 1). He calls thiswarm/moist climatic phase the mesothermic period.Temperatures apparently reached modem levels inthe southern interior of British Columbia and northeast Washington around 4500 to 2500 14C yr BP (Alley, 1976; Mack et at, 1978a,b; Mathewes and King,1989; Hebda, 1995). This late Holocene climaticdeterioration correlates with neoglacial advances inthe Coast and Rocky Mountains (Porter and Den-ton, 1967; Ryder and Thomson, 1986; Clague, 1989;Luckman et at, 1993; Mann and Hamilton, 1995).

Subalpine treelines are climatically sensitive tension zones, and are therefore ideal locations forreconstructing palaeoclimatic regimes (Clague andMathewes, 1989; Pellatt and Mathewes, 1994). Inthis study we employ palynological analyses, as wellas chironomid head capsules from a sediment corefrom Cabin Lake, British Columbia (121°13.2’W,49°58.4’N), to reconstruct the Holocene historyof vegetation and climate in the EngelmannSpruce—Subalpine Fir biogeoclimatic zone. These

analyses are supported by pollen ratio analysis,tephrochronology, AMS radiometric dating, loss onignition, and statistical zonation, and allow us tocompare the timing of climatic changes betweenhigh and low elevations in the interior, and betweenhigh elevation sites on the coast and in the interior.

2. Study area

2.]. Physiography

Cabin Lake is located at 1850 m asl on StoyomaMountain (2283 m asl; 121°13’W, 49°59’N) in thesouthwestern interior of British Columbia (Fig. 2)at the northern limit of the Cascade Mountains. TheCascade Mountains of British Columbia merge intothe Kamloops Plateau to the east, and are separatedfrom the Coast Mountains to the west by the FraserRiver (Holland, 1976). The eastern margin of theCascade Mountains is a transition zone where summit elevation and dissection progressively decreasetoward the Kamloops Plateau (Holland, 1976). Thepeaks of the Hozameen Range of the Cascade Mountains are characterized by high, serrated ridges thatshow the effects of intense alpine glaciation (Holland, 1976). Cirque basins are common on north-and northeast-facing slopes of peaks and ridges.At lower elevations, between 1830 and 2135 m,there are rounded ridges and dome-shaped mountains which were over-ridden by ice at the maximumof the Cordilleran ice sheet (Holland, 1976).

2.2. Vegetation and climate

Stoyoma Mountain is located in the central dryclimate region of the Kamloops Forest Region(Lloyd et al., 1990). The ESSF zone is the uppermostforested zone in the southern three-quarters of interior British Columbia (Coupe, 1983; Meidinger andPojar, 1991). The continental climate is relativelycold, moist, and snowy, with a short growing season,and long, cold winters. Mean annual temperature is—2 to 2°C with 5 to 7 months below 0°C and only 2months or less above 10°C (Coupe, 1983; Meidingerand Pojar, 1991). Precipitation ranges from 400 mmin the drier portions to 2200 mm in the wetter areas.As much as 70% of the precipitation falls as snow.

M.G. Fe/tart et aL /Palaeogeography, Falaeoclimatology, Palaeoeco!ogy 141 (1998) 123—138 125

Engelmann spruce (Picea engelmannii) and subalpine fir (Abies lasiocarpa) are the climax treesof the ESSF. Engelmann spruce is longer-lived andthus often dominates the canopy of mature standswith subalpine fir in the understory. Subalpine fir becomes dominant at the higher elevations of the ESSFand in wetter areas (Meidinger and Pojar, 1991).Whitebark pine (Finns albicaulis) appears in drierparts of the ESSF and mountain hemlock (Tsugamertensiana) may occur in the western portion of

COOLMOIST

the ESSF near the Mountain Hemlock zone. TheCabin Lake study site is located in the dry, coldEngelmann Spruce—Subalpine Fir subzone variant 2(ESSFdc2) (Lloyd et al., 1990). Typical trees andshrubs encountered are subalpine fir, Engelmannspruce, whitebark pine, black huckleberry (Vaccinium membranaceum), and white-flowered rhododendron (Rhododendron alb(florum). Typical dwarfshrubs, herbs and mosses include grouseberry (Vaccinium scoparium), five-leaved bramble (Rubus pe

“C yr BP

0 —,

NORTHEASTERN SOUTHERN FRASER LOWLAND IDF- Marion Lake, LOLIIS Pond.WASHINGTON INTERIOR B.C. & CANYON (Mathewes SouttMe% B.C. Queen charlotte lelands, B.C.

(Melvinger. 1985) (Hobda, 1998) (Mathewes & Rouse. 1975) & Kng,I989) (Mathewes, 1973) (Pellatt & Mathewes. 1994)

~ MODERN MODERN

MODERN2

3-

4—

5

6—

7—

a—

9-

MESOTHERMIC

WARM,MOIST

MODERN

NEARMODERN

EARLYHOLOCENEThERMALMAXIMUM

WARM,DRY

MODERN

EARLYHOLOCENETHERMALMAXIMUM

WARM,DRY

NEARMODERN

EARLYHOLOCENETHERMAl.MAX MUM

WARM,DRY

EARLYHOLOCENEThERMAl.MAX MUM

WARM,DRY

NEARMODERN

WARM,MOIST

EARLYHOLOCENETHERMALMAXiMUM

WARM,DRY

11 —COOLMOIST

12 —

COOL

13 —

COLD

Fig. 1. Comparison of climate change at selected sites11W = Interior Douglas Fir Zone.

along the Pacific Northwest coast and in the southern interior of British Columbia.

126 M. G. Pdllat: et a!. /Palaeogeography, Palaeoclimatcilogy, Palaeoeco!ogy 141 (1998) 123—138

c/titus), mountain arnica (Amica lauifolia), Sitka Valeiian (Valeriana sitchensis), and red-stemmed feathermoss (P/euroziwn schreberi) (Lloyd et al., 1990).

2..3. Study site:

• Cabin Lake (Figs. 2 and 3) is locatçd at 1850 m• •asl; It is about 4 ha in area with a maximum water

depth of 4.2 m. A small intermittent inlet stream runsinto the north end of the lake. It carried water in

July but was dry in late August of .1995. An overflowoutlet drains the south end of the lake when waterlevels permit. ,.

A well developed forest surrounds Cabin Lake onthe east, south, and part of the west slopes. A firehas burned much of the slope north of the lake leav~..ing open meadow vegetation. Some bf the commontrees and shrubs surrounding Cabin Lake are Engelmann spruce (the dominant tree species), subalpinefir, white-flowered rhododendron, Luetkea pectinata,

BRITISH COLUMBIA

QUEENcHAR LOTTE

ISLAflDS

PACIFICOCEAN

Fig. 2. Map of British Columbia indicating the location of the Cabin Lake study site.

490..

MG. Pel!au et aL !Palaeogeography, Pa!aeoclimatology. Pataeoecology 141 (1998) 123—138 127

Vaccinium membranaceum, Vaccinium scoparium,Phyllodoce empe:rjformis, and Cassiope mertensiana. Some of the common herbs include ValeHants sitchensis, Veronica cf. worm.skjoldii, Castillejaminiata, Saxifraga ferruginea, Senecio triangularis,Leptarrhena pyrolzfolia, Lupinus arcticus, Calthaleptosepala, Arnica cf. cordifolia, Anemone occidentalis, Caltha biflora, Pyrola sp., Carex sp., andEriophorum sp.

3. Methods

3.1. Pollen analysis

A 399-cm sediment core was recovered near thecentre of Cabin Lake at a water depth of 4.2 m.

Subsamples of 1 ml were removed at 5 cm intervals,except in the basal 91 cm of clay and silt where2 cm intervals were processed for pollen. Volumeswere determined by displacement in water, using a10 ml graduated cylinder. A known concentrationof marker spores (11, 300 ± 400 Lycopodium) wasadded to the subsamples before processing. The protocols for pollen extraction follow Berglund and Ralska-Jasiewiczowa (1986). Identifications of pollenand spores were aided by published keys (MeAndrews et al., 1973; F~gri and Iversen, 1989; Mooreet al., 1991) and the Simon Fraser University modem reference collection. Routine counting of palynomorphs was carried out at 500 x magnificationand critical identifications were made under oil immersion at 1200x. The basic pollen sum (between500 and 875 grains) used for percentage calculationincludes all terrestrial pollen.

Raw data were converted into percentages usingTILIA 2.0 (Grimm, 1993). TILIAGRAPH 1.25 wasused to generate the pollen diagrams (Figs. 4 and 5)which were subdivided into local pollen zones usingconstrained cluster analysis. The computer programused for statistical zonation was CONISS (Grimm,1987). Plant taxa used to generate the zonation dendrogram included trees and shrubs with values of atleast 2% in two intervals. Mount Mazama (6730 14Cyr BP) and Bridge River (2435 ‘4C yr BP) tephrasare present in the Cabin Lake sediment core. Thesewell-dated tephras (Clague et al., 1995; Hallett et al.,1997) provide chronological control, along with 14Cdating. Tephras were identified by Gerald Osbornand Glen DePaoli at the University of Calgary usingmicroprobe analysis. In order to supplement the agecontrol provided by the tephras, three AMS radiocarbon ages were also determined (Table 1). Oneage is based on dating of a pollen concentrate, prepared by the method of Brown (1994). Radiocarbonages between dated levels were interpolated usingpolynomial regression analysis (Grimm, 1993).

It is well known that Diploxylon pine (Pinuscontorta type) pollen is greatly overrepresented inpollen assemblages from the ESSF (Hebda, 1995;Pellatt, 1996; Pellatt et al., 1997). This pollen islargely of regional and extra-local origin and doesnot represent the local vegetation at Cabin Lake.In order to increase the resolution of local pollentypes, a~pollen diagram with lodgepole pine removed

—~ dna

Pig. 3. (a) Photograph of Cabin Lake showing ESSF forest surrounding the lake. (b) Photograph of Stoyoma Mountain showingthe transition from ESSE forest to alpine tundra.

Co

c4

Z4~5o2S —

Moon Tooko€730040

o,I0.iOO.

l0.000070

J~:~1 ~?1it

Gylijo Cloy SIll

Fig. 4. Percentage pollen diagram for Cabin Lake, with lOx exaggeration curves (stippled) to highlight abundances of infrequent pollen types. AMS datçs are shown on the ~left. Pollen concentrations are x 100. Zones were derived by stratigraphically constrained cluster analysis (CONISS).

NJ

to.,00

It

ItIt

~ TipIlil243526

4.!~SI3G.ID

a~Io.I~o_

O.CSO•7G

9O~O’6o

Fig. 5. Percentage pollen diagram with lodgepole pine removed.for Cabin Lake, with lOx exaggeration curves (stippled) to highlight abundances of infrequent pollen types.AMS dates are shown on the left. Pollen concentrations are x 100. Zones were derived by stratigraphically constrained cluster analysis (CONISS).

p.rttctl poIt~n

Gylijo Cloy Silt

~0

130 M. G. Pellan et a?. /Pa?aeogeography, Palacoclimatology, Palaeoecology 141 (1998) 123—138

Table 1AMS radiocarbon dates for the Cabin Lake sediment core

Sample identification Sample description Lab No. Age, ‘4C yr BP

CLJL94-5-12, 304cm carbonized wood sample T0-5205 8,910 ± 120CLJL94-S, 324.cm pollen CAMS-29826 10, 090 ± 70CLJL94—5, 368—370 cm leaf fragment CAMS-29829 9, 860 ± 60

Errors presented at ±1 s.

—

~_____

Fig. 7. Spruce/Diploxylon pine percent pollen ratios for surfacesediment samples at selected elevations in the ESSE zone.

from the pollen suni was also prepared (Fit. 5).Spruce/Diploxylon pine percent ratios• (spruce/pineratios) were calculated (Fig. 6) to assist with interpretation-of changes. These ratios are compared withthe spruce/jine ratios (Fig. 7) calculated for modemlake surface sediment samples (Pellatt, 1 996~ Pellattet a, 1997).

3.2. Chironomid analysis

Cabin. Lake sediments were mostly subsamplede~’ery 15 cm, with higher resolution sampling inregions of expected faunal change (i.e., suspected

late-glacial basal clay and basal àlay/gyttja interface). Subsamples normally consisted of 0.5 ml orI ml of sediment, but up to 10 ml of sedimentwas necessary in some intervals to obtain sufficientnumbers (at least 30) of chironomid head capsulesfor analysis. Isolation of -chironomid head capsules,Chaoborus mandibles, - and ceratopogonid head cap~sales followed the procedures outlined by Walker(1987).

Remains were identified at 100—400x magnification. Identifications were based on descriptionsand keys by Oliver and Roussel (1983); Wiederholin (1983) and Walker (1988). Whole head capsules and fragments containing greater than halfof the mentum were counted as one head capsule. Fragments- that were exactly half of a headcapsule were counted as one half, and fragmentsthat consisted of less than half of the mentumwere not counted. Most identificatiàns were madeat the generic level, although a few species identifications were possible. Broader taxonomic categorieswere necessary where genera could not be determined (i:e., Tanytarsina, Cricotopus/Onhocladius,Coiynoneura/Thienemanniella). Data were compiled on a spreadsheet using TILL4 2.0 (Grimm,1993), ~nd chironomid percentage diagrams werepioduced using TILIAGRAPH 1.25 (Grimm, 1991)(Fig. 8). A constrained sum-of-squares cluster analysis (CONISS) was done to examine major changesin chironomid communities (Grimm, 1987).

3.3. Loss on ignition

I ml of sediment was sampled at 5 cm intervals todetermine organic content through loss on ignition.Protocol for the procedure was taken from Hakansonand Jansson (1983) in which dried sediment wascombusted at 550°C for 1 h. Percent LOl results areshown in Fig. 4 along with the pollen data.

0 50 100 150 200 250

Depth (cm)

300 350 400

Fig. 6. Spruce/Diploxylon pine percent pollen ratios for CabinLake.

2210

-

2060

2030

2006

!2000

.0 1960l~70

~1~0

0 0.2 0.4 0.6- 0.6 12

Ratio —

M.G. Pellat: at aL /Palaeogeography, Palaeoclimatotogy, Pa!aeoecolo~ 141 (1998) 123—138 131

M,z~ta Ass,673040

9910a120_

10.09000

9850160-

4. Results and discussion

4.1. Pollen assemblage zones at Cabin Lake

The Cabin Lake pollen assemblages are shownin Fig. 4. A pollen diagram with Diploxylon pineremoved from the pollen sum is shown in Fig. 5.Spruce/pine ratios are plotted in Fig. 6.

4.1.1. Zone CP-1 (Pinus cf contofla — Picea— Poaceae — Aflemisia, 399—312 cm, late-glacial,>10,090 ± 70 ‘4CyrBP)

In Zone CP-1, lodgepole pine type (Diploxylonpine) pollen percentages exceed 80%, the highestlevels recorded in the core. Spruce pollen is alsoan important component of the assemblage, pealcingat 350 cm along with Silica alder and an abundance of herb pollen. Herbs such as grass (Poaceae),sedge (Cyperaceae), Artemisia, Caryophyllaceae,Epilobium, Asteraceae, Selaginella densa type, andBotrychium reach their highest levels here, and indicate an open subalpine/alpine vegetation covetThe sediments in this zone are composed of clay

with sand/silt bands (Figs. 4 and 5). Total pollenconcentration was extremely low throughout, anddramatically increases at the clay/gyttja interface(324 cm; 10,090 + 70 ‘4C yr UP). An AMS dateof 9860 ± 60 was determined from a leaf fragmentbetween 368 and 370 cm. Variability in ‘4C production, commonly referred to as a ‘4C plateau, causesmultiple calibrated ages for ‘4C over certain timeperiods (Stuiver et al., 1991; Bartlein et al., 1995).Therefore, this 9860 ± 60 ‘4C yr BP age is not considered significantly different than the 10, 090 ± 70

yr UP age obtained at 324 cm, at the clay/gyttjainterface (Table 1).

Low pollen concentration in conjunction with lowspruce/pine ratios indicates an open vegetation coverand/or rapid sedimentation. Pine and spruce are thedominant pollen types, even though this zone probably represents an alpine tundra-like environmentwith only scattered spruce krumxnholz. Macrofossils obtained from silt bands in this zone includeElaeagnaceae trichomes, dwarf willow (Salix) leaffragments and buds, a cf. Dryas leaf fragment, and aPhyllodoce needle. These macrofossils are indicative

c4 ‘Foi

50

5,id9. R~ar T.ptvo243526 Ice.

150-

aoo

25D

300

350•

400

Gytijo Cloy Silt

Fig. 8. Chironoinid head capsule percentage diagram for Cabin Lake. AMS dates are shown on the left. Zones were derived bystratigraphically constrained cluster analysis (CONISS).

C

132 M.G. Peitate at at. /Palaeogeography, Pataeoctinwtology, Pataeoecotogy 141 (1998) 123—138.1

of a continental alpine/subalpine plant community.Thus late-glacial pollen and plant macrofossils suggest cold conditions at Cabin Lake.

4.1.2. Zone CP-2 (Pinus cf contorta — Picea—Alnus viridis, 312—2 70 cm, 10,090 + 70 to Ca.7000 ‘4CyrBP)

In Zone CP-2, lodgepole pine type pollen decreases relative to CP- 1, but remains at over 40%of the pollen sum. Sitka alder type pollen achievesits highest levels in the core. Spruce and Abiespollen increase throughout this zone and whitebarkpine type (Haploxylon pine) pollen is fairly high.Anemisia and Poaceae are the dominant herbs. Valeriana sitchensis and Liliaceae enter the pollen recordin this zone. Total pollen concentration is at its highest here, attaining values of over 80,000 grains perml. AMS radiocarbon dates of 10, 090 + 70 wereobtained on a pollen concentrate from 324 cm and,8910 + 120 14C yr BP on a piece of wood at 304cm(see Table I).

Relatively high levels of spruce and Abies pollensuggest that trees typical of the ESSF zone werepresent during this early Holocene period. The highvalues of lodgepole pine (-~-‘40 to 50%) and Sitkaalder pollen are most likely of regional and extra-local origin and suggest open conditions at CabinLake. High levels of spruce and Abies, in conjunctionwith initially high levels of Sitka alder, suggest thatan environment with no modem analogue existed inthe early Holocene. The absence of cool or moistindicators such as Ericales (heaths and Empetrum)and the significant presence of shade-intolerant taxalike whitebark pine type, Poaceae and Artemisia suggest a city climate with open growing conditions.Relatively high spruce (probably Picea engelmannii) and Haploxylon pine (probably whitebark pine)pollen values, and low Abies pollen values (probablysubalpine fir) suggest warmer/drier conditions thanpresent. This interpretation is consistent with theearly Holocene xerothermic period noted in coastalBritish Columbia (Mathewes, 1973; Mathewes andHeusser, 1981; Hebda, 1995).

4.1.3. Zone CP-3 (Pinus cf contorta — Picea— Abies, 270—190 cm, ca. 7000—4800 ‘4C yr BP)

In Zone CP-3, levels of lodgepole pine typeand spruce pollen remain relatively constant. Abies

pollen increases relative to the previous zonewhereas whitebark pine (Haploxylon) type and Sitkaalder decrease. Aquatic Isoetes microspores appearfor the first time.

This zone appears to represent a period of vegetational and climatic transition at Cabin Lake. Abies(probably subalpine fir) becomes the dominant treearound Cabin Lake. Because Abies is under-represented in modem pollen assemblages (Dunwiddie, 1987; Hebda and Allen, 1993), 20% abundancesuggests that Abies dominated the surrounding forest (Hebda and Allen, 1993). High Abies valuesand other subalpine indicators of moisture, such asCyperaceae and Ericales, indicate that conditionswere wetter than inferred in either zones CP-1 orCP-2. High pollen concentrations (Figs. 4 and 5)in conjunction with high organic content (% LOT)and low spruce/pine ratios (Fig. 6) indicate thatclimate was wanner than present. The increases inthese subalpine taxa indicate that temperature wasbeginning to decrease during this zone. This zonecorresponds with the mesothermic period observedin the southern interior of coastal British Columbia(Hebda, 1995).

4.1.4. Zone CP-4a (Pinus cf contofla — Picea— Tsuga heterophylla — Cyperaceae — Ericales,190—95 cm, 4800—2435 14C yr BP)

In Zone CP-4a lodgepole pinft type pollen decreases to about 40% of the pollen sum. Sprucepollen increases and Abies decreases from CP-3 butremains as an important component of the pollensum. Cupressaceae, Ericales, Rosaceae, and Cyperaceae all increase in abundance. Caitha b(flora andRanunculus enter the pollen record and Isoetes remains important. Regional transport of western hemlock pollen (Tsuga heterophylla) increases.

The increased values of Ericales, Caitha b(flora,Ranunculus type, Rosaceae, and Cyperaceae pollenindicate that typical subalpine vegetation had become established. A significant decrease in pollenconcentration occurs — lower concentrations wereonly recorded in zone CP-1 (late-glacial). Values ofregionally transported western hemlock pollen attain theft highest values in this zone, suggesting adecrease in local pollen productivity, decreased temperature, and increased precipitation at lower elevations, as observed in the Fraser Canyon (Mathewes

M.G. Pellati et al. /Pa!aeogeography, Palaeoclimatology, PalaeoecoIo~ 241 (1998) 123—138 133

and King, 1989). It appears that climate was coolerthan in zones CP-2 and CP-3, with increased importance of Engelmann spruce in the surroundingforest.

4.1.5. Zone CP-4b (Pinus cf contorta — Picea— Abies, 95—65 cm, 2435 to ca. 1700 ‘4C yr BP)

• In CP-4b Abies, lodgepole pine type, Poaceae andCyperaceae pollen increase whereas spruce pollendecreases. Subalpine/alpine herbs remain diverseand pollen concentrations remain low. Bridge Rivertephra (2435 ‘4C yr BP) occurs at the base of thiszone.

Increases in Abies and Cyperaceae indicate thatconditions may have been wetter than those observed in CP-4a, but remaining cool. This increasedwetness, corresponding with cool/wet conditions after deposition of the Bridge River tephra (ca. 2435

yr BP), may be the same climatic factor thatpromoted glacial advances in the northern CascadeMountains (moraines of the Burroughs MountainStade 2050 ‘4C yr BP), and the Canadian RockyMountains (Porter and Denton, 1967; Luckman etal., 1993), as well as cool/moist conditions in the Intenor Douglas Fir (IDF) zone of the Fraser Canyon(Mathewes and King, 1989).

Cool/moist conditions in CP-4b correspond withthe development of modern forests in the FraserCanyon (Mathewes and King, 1989), and in the interior Pacific Northwest, U.S.A. (Mehringer, 1985).At the same time pollen ratio analysis suggests thatlocal spruce production was low, indicating that theenvironment was likely more open than in zoneCP-4a (Fig. 6).

4.1.6. Zone CP-5 (Pinus cf contona — Picea— Pinus cf albicaulis, 65—0 cm, Ca. 1700 14C yr BPto present)

In Zone CP-5 whitebark pine type, spruce, andIsoetes pollen and spores increase. Abies, Sitka alder,and Ericales decrease. Pollen concentration increasesto levels not seen since CP-3 and CP-2. Diversity ofsubalpine/alpine herbs decreases. An increase inspruce/pine ratio values is observed (Fig. 6).

This zone represents modern conditions at CabinLalce. High values of whitebark pine and Engelmann spruce indicate that conditions are drier thanin Zones 4a and 4b. This relatively drier, possibly

wanner phase appears to correspond with conditionsin the interior Pacific Northwest, U.S.A. (Mehringer,1985), but still represents cooler conditions thanobserved in the early Holocene xerothermic andmesothermic periods. -

4.2. Chironomid assemblage zones at Cabin Lake

The chironomid fossil record of Cabin Lake hasbeen divided into four zones representative of themain changes that occur in chironomid communitycomposition (Fig. 8). Zone 1 (399—330 cm) represents Late-Pleistocene sediments before ca. 10,000yr BP at the clay/gyttja transition. Zone 2 (330—265 cm) encompasses the early Holocene, between10,000 yr BP and ca. 7200 ‘4C yr BR Zone 3 (265—112 cm) spans the mid-Holocene between ca. 7200and ca. 3200 14C yr BR The most recent sedimentsmake up Zone 4 (112—0 cm), representing ca. 3200

yr BP to present.

4.2.1. Zone CC-I (399—330 cm, late-glacial,>10,090±70 ‘4CyrBP)

The late-glacial assemblage primarily consists ofthe widespread Tanytarsina group (up to 54%), andcold-stenotherms typical of oligotrophic (Paracladius, Parakiefferiella nigra, Mesocricotopus, Stictochironomus, Protanypus, and Heterotrissocladius),and mesoirophic (Sergentia) waters. The predatoryProcladius also makes up a significant proportion ofthe late-glacial assemblage (up to 33%), and remainsrelatively abundant throughout the Holocene.

The late-glacial chironomid assemblage at CabinLake is very similar to the ‘late-glacial Heterotrissocladius fauna’, coined by Walker and Mathewes(1987a) for coastal British Columbia sites, and is alsofound in New Brunswick (Levesque et al., 1993) andGermany (Hofmann, 1983). This assemblage consists of typical cold-stenothermous taxa whose distributions are primarily restricted to cold oligotrophicarctic and alpine waters, or the deep, cold profundalregions of large, thermally stratified temperate lakes(Walker, 1987, 1990; Walker and Mathewes, l987a,1989a). The Cabin Lake sediments show no indication of thermal stratification since this relatively shallow lake was formed, and so it is reasonable to assume that there was no deep, cold habitat isolatedfrom the influences of climate at the surface waters.

t

134 M.G. Pe!latt et a!. /Pa!aeogeo~raphy, Palaeoclinwtology, Palaeoeco!agy 141 (1998)123—138

Thus, the cold-stenothermous assemblage suggeststhat the late-glacial water temperature, and presumably air temperature, was relatively cold. This is consistent with conclusions based on pollen and chironomid inferences on the British Columbia coast (Math-ewes, 1973; Mathewes and Heusser, 1981; Walkerand Mathewes, 1987a, 1989a).

4.2.2. Zone CC-2 (330—265 cm, 10,090 ± 70 to ca.7200’4CyrBP)

The beginning of the Holocene is characterized bythe sudden disappearance of most cold-stenotherms,with only Heterotrissocladius and Sergentia persisting as small proportions of the faunal assemblage. Inconjunction with this trend, significant increases in

- the warm-adapted chironomids occur. Most notably,Chironomus, Stempellinella/Zavrelia, Pagastiella,and Microtendipes, along with Chaoborus, rapidlybecome major faunal elements in the early Holocene.The eurythennic Tanytarsina and Procladius approach their lowest levels in this zone, which may bea result of less production in these groups or simplybecause of a greater prevalence of other taxa. Thiszone also shows an increase in proportions of therheophilous Corynoneura/Thienemanniella group,

- and the eurythennic Psectrocladius and Corynocera.The rapid decline in cold-stenothermous taxa,

and subsequent dominance of typical warm-watertàxa strongly suggest significant climatic warm-

-- ing. Although not as stenothermic as the cold-watertaxa, the warm-adapted chironomids in this assem

• blag~ are most often found in warm, temperate waters (Walker, 1987; Walker and Mathewes, 1987a,b;1989a,b; Walker et a, 1997). In .the relatively uncommon instances where a few of these taxa occur inarctic or alpine sites, they are restricted to small shallow ponds which attain high summer water temperatures (Walker and MacDonald, 1995). This warmearly Holocene period at Cabin Lake cónfinns thewidespread extent of the early Holocene xerothermic interval as describedby Mathewes and Heusser

• (1981), Mathewes (1985); Hebda (1995) and Elias(1996).

4.2.3. Zone CC-3 (265—112 cm, ca. 7200 to Ca. 3200‘4Cyr BP)

Just prior to Mazama ash deposition, a major shiftin fauna occurs. Although this mid-Holocene zone

still supports a large group of warm-adapted chironomids, a notable reduction in Stempellinella, Zavreliaand Pagastiella is seen, and Sergentia, a cold-watergenus, reaches significant levels (up to 27%). Reterotrissocladius also increases to levels comparableto its late-glacial abundance. Chironomus remalnsa dominant genus through this zone (up to 27%),with Microtendipes consistently making up approximately 5% of the community. Other important warm-water taxa comprising this zone are Dicrotendipesand Parakiefferiella cf. bathophila, with Chaoborusmandibles reaching theft greatest abundance. Tanytarsina (37%). and Procladius (34%) again attainsignificant relative abundances in the mid-Holocene,and various rheophilous taxa continue to. be consistently represented in small proportions.

The decline in some wann-water chironomids(most notably Stempellinella, Zavrelia and Pagastiella) and the significance of the cold-stenothermic Sergentia suggest that this period was characterized by a cooling trend. Although Sergentiais indicative of cold waters (Walker et al.,- 1997),the persistence of a relatively diverse and abundantwarm-water assemblage reveals that water temperature had cooled in comparison to the early Holocene,but was still wanner than present. In addition Sergentia is commonly found in waters with moderateoxygen depletion, often in shallow lakes that freezeto the bottom in winter, as Cabin Lake may have inthe mid-Holocene. Increased precipitation may haveplayed a role in winter anoxia. Greater snowpackon the lake would lead to longer winter conditions,•as the time required for the snow to melt wouldbe extended and anoxic conditions would persist•for a longer period. This transitional cooling trendthroughout the mid-Holocene at Cabin Lake supportsHebda’s (1995) designation of a mesothermic periodin the southern interior of B.C.

4.2.4. Zone CC-4 (112—0 cm, 3200 ‘4CyrBP to- present)

The most recent, late-Holocene sediments reveala dramatic decrease in the relative contribution ofwarm-water taxa as a group, with many genera becoming locally extirpated and others reaching theirminimum Holocene values. Sergentia remains relatively abundant, with Heterotrissocladius re-attaining its typical late-glacial abundance at the begin-

M.G. Pellatt et aL /Palaeogeography, Palaeoclimatology, Pakicoecotogy 141 (1998) 123—138 135

ning and end of this interval. Possibly significantis the reappearance of Paracladius, a strongly coldstenothermous taxon (Walker et al., 1997), in thelate-Holocene fossil assemblage.

Procladjus remains a dominant faunal element,whereas Tanytarsina reaches its maximum abundance (61%), surpassing its late-glacial abundance.Rheophilous taxa continue to be deposited into thelake, although a less diverse group is evident in thelate-Holocene.

The persistence of Sergentia and Heterotrissocladius, and further reduction in the warm-waterchironomid assemblage in the late Holocene indicate further cooling. This is consistent with observedcooling in the southern interior of British Columbia(Mathewes and King, 1989) and corresponds to thetiming of neoglacial advances in the northern Cascades (Porter and Denton, 1967).

5. Conclusions

Four main periods of climate are documentedsince deglaciation at high elevations in the Canathan Cascade Mountains. These periods are: (1) alate-glacial cold period (>10,090 14C yr BP); (2) anearly Holocene wann, dry period (10,090 to 7000l4~ yr BP); (3) a mid-Holocene period of climatic

transition beginning with a warm, moist phase from7000 to 4800 ‘4C yr BP and then cooling between4800 to 2435 14C yr BP and; (4) modern neoglacial(cool/moist) conditions (2435 ‘4C yr BP to present).

The oldest palaeobotanical evidence from CabinLake indicates continental, alpine tundra conditionsin the late-glacial. Elaeagnaceae trichomes (probablyShepherdia canadensis), cf. Dryas leaf fragments,dwarf willow remains and a Phyllodoce needle attest to a cold continental climate during this time.Abundant Shepherdia pollen was reported by Clagueet al. (1995) from early Holocene peat above thepresent treeline in the Coast Mountains. Chironomidcommunities in Cabin Lake during the late-glacialconsist of typical cold-adapted species.

Early Holocene conditions at Cabin Lake maywell represent a vegetation assemblage that has nomodem analogue. High values of lodgepole pinetype, spruce, and Sitka alder type pollen indicateopen parkland conditions. This spruce—fir parkiand

environment suggests that day conditions hinderedthe development of closed forests. This data, in conjunction with a warm-adapted chironomid assemblage indicates that the early Holocene was warmand dry at Cabin Lake and corresponds with theearly Holocene xerothermic period.

-After ca. 7000 14C yr EP, percentages of spruceand Abies pollen increase. This palaeobotanical evidence implies that ESSF forests developed as precipitation increased toward the end of the earlyHolocene xerothermic period (Mathewes and King,1989; Hebda, 1995). Chironomid evidence suggestsa slight cooling and possibly increased precipitationduring the mid-Holocene. The changes in pollen andchironomid assemblages between ca. 7200 and ca.7000 ‘4C yr BP appear to follow a notable cooling event detected from Greenland, Antarctica, andAfrica around 7500 14C yr BP (Alley et al., 1997;Stager and Mayewski, 1997). This cooling event wasapproximately half the amplitude of the YoungerDryas and may well have had a global expression(Alley et al., 1997).

Hebda (1995) suggests that a climatic periodtermed the ‘mesothermic’ be incorporated into theclimatic history of British Columbia. He suggeststhat this period may be an extension of the Hypsithermal, extending thQ timing of Holocene maximum wannthtoaround 4000 14C yr BP. This wouldincorporate the timing of the Hypsithermal in mostof Canada (Anderson et al., 1989). Palaeobotanicalevidence at Cabin Lake indicates that early Holocenewarmth extended into the mid-Holocene, thus supporting Hebda’s hypothesis. Warm, moist conditions,relative to present climate, led to the developmentof ESSF forest at Cabin Lake. A modem chironomid community became established at Cabin Lakearound 3200 ‘4C yr BP whereas modern pollen assemblages appear at Cabin Lake after deposition ofthe Bridge River tephra (2435 ‘4C yr EP). A period of increased moisture falls between 2435 andca. 1700 ‘4C yr BP, corresponding with neoglacialadvances throughout the Canadian Cordillera.

Although the pattern of change between pollenand chironomid assemblages at Cabin Lake is similar, it is important to note that there is a definitelag in the response time between chironomids andvegetation (Fig. 9). Time lags between plant andinsect migration and establishment have been ob

t

136 M.G. Peilau et aL !Palaeogeography, Pa!aeoclimatology. Palacoecology 141 (1998)123—138

300

400

Fig. 9. Comparison between pollen and chironomid assemblagezones for Cabin Lake.

served at study sites in eastern North America and inEurope (Elias, 1994) and are generally attributed tothe slow regeneration time of arboreal communities(Brubaker, 1986) and the rapid dispersal ability offlying insects. This may be true at Cabin Lake aswell, although differences in the sensitivity of chironomids and vegetation to climate change may alsoaccount for the observed time lags. Further multi-proxy studies are needed to determine the responsetimes of community change in vegetation and chironomids to climate change. As a whole, this studyhas served to illustrate the importance of multiproxypalaeoecological investigations at ecologically sensitive boundaries such as treeline.

Acknowledgements

We would like to thank the Natural Sciences andEngineering Research Council (NSERC) of Canadafor their support in funding this project via grants to

References

Alley, NE, 1976. The palynology and paleodimatic significanceof a dated core of Holocene peat, Okanagan Valley, southernBritish Columbia. Can. J. Earth Sci. 13, 1131—1144.

Alley, RB., Mayewski, P.A., Sowers, T., Stuiver, M., Taylor,ICC., Clark, EU., 1997. Holocene climatic instability: Aprominent, widespread event 8200 yr ago. Geology 25, 483—486.

Anderson, T.W., Mathewes, R.W., Schweger, CE., 1989.Holocene climatic trends in Canada with special referenceto the Hypsithermal interval. In: Fulton, R.3. (Ed.), Quaternary Geology of Canada and Greenland. Geological Survey ofCanada, Geology of Canada 1, pp. 520—528.

Bamosky, C.W., Anderson, P.M., Bartlein, P.3., 1987. The Northwestern United States during declaciation: vegetation andclimatic implication. In: Ruddiman, WE, Wright, HE., Jr.(Eds.), North America and adjacent oceans during the lastdeglaciation. Geology of North America Vol. K-3, pp. 289—321.

Bartlein, P3., Edwards, M.E., Shafer, S.L., Barker Jr, ED.,1995. Calibration of radiocarbon ages and the interpretation ofpaleoenvironmental records. Quat. Res. 44, 417—424.

Berglund, B., Ralska-Jasiewiczowa, M., 1986. Pollen analysisand pollen diagrams. In: Bergland, B. (Ed.), Handbook ofPalaeoecology and Palaeohydrology. Wiley, Torontq, Ont, pp.455—484.

Brown, T.A., 1994. Radiocarbon Dating of Pollen by Accelerator Mass Spectrometty. Ph.D. Thesis, Univ. of Washington,Seattle, Wash., 286 pp.

Brubaker, L.B., 1986. Responses of tree populations to climaticchange. Vegetation 68, 119—130.

Clague, 3.3., 1989. Introduction (Quaternary geology of theCanadian Cordillera). In: Fulton, Ri. (Ed.), Quaternasy Geology of Canada and Greenland. Geological Survey of Canada,Geology of Canada 1, pp. 19—22.

Clague, 3.3., Mathewes, R.W., 1989. Early Holocene thermalmaximum in western North America: New evidence from.Castle Peak, British Columbia. Geology 17, 277—280.

Clague, J.J., Evens, S.G., Rampton, V.N., Woodsworth, G.J.,1995. Improved age estimates for the White River and BridgeRiver tephras, western Canada. Can. J. Earth Sd. 32, 1172—1179.

Coupe. R., 1983. Engelmann Spruce—Subalpine Fir Zone. In:Watts, S.B. (Ed.), Forestry Handbook for British Columbia.Faculty of Forestry, University of British Columbia, vancouver, B.C., pp. 273—277.

Dunwiddie, P., 1987. Macrofossil and pollen representation of

RWM and tRW. We would also like to thank MarkusHeinrichs, Doug Cheetham, and André Tessier fortheir help in the field. Our appreciation is extended toS.A. Elias and an anonymous reviewer for commentsthat improved the manuscript.

Pollen Zones chnonomid Zones

0-i

I GP-5

r700’4cyrBP

I CP-4b2410’tyrBP

I 3210’CyrBPCP-4a

I ___ cc-a4800 It yr BP

200—H

CP-3

7000”CyrBP 722o”cyrBP

CP-2 cc-2

10.090 ‘t yr EP 10,090 ‘4c yr BP

cP-1 cc-i

kf.G. Peltatt et aL /Paiaeogeography, Palaeoclima:ology, Palaeoeco!ogy 141 (1998) 123—138 137

coniferous trees in modem sediments from Washington. Ecology 68, 1—11.

Elias, S.A., 1994. Quaternary Insects and their Environments.Smithsonian Institute Press, Washington, D.C., 284 pp.

Elias, S.A., 1996. Late Pleistocene and Holocene seasonal temperatures reconstnacted from fossil beetle assemblages in theRocky Mountains. Quat. Res. 46, 311—318.

&gri, K., Iversen, 3., 1989. Textbook of Pollen Analysis (4thed.). Wiley, London, 328 pp.

Fedje, D.W., 1993. Sea-levels and Prehistory in Gwaii Haanas.MA. Thesis, Univ. Calgary.

Grimm, E., 1987. CONISS: a FORTRAN 77 program for strati-graphically constrained cluster analysis by the method of incremental sum of squares. Comput. Geosci. 13, 13—35.

Grimm, E., 1991. TILIAGRAPH vl.25 (computer software). 1111-nois State Museum, Research and Collections Centre, Springfield, Ill.

Grimm, E., 1993. TILLk v2.0 (cothputer software). illinois StateMuseum, Research and Collections Centre, Springfield, flI.

Hakanson, L., Jansson, M., 1983. Principles of Lake Sedinientology. Springer, New York, N.Y., 316 pp. -

Hallett, D.J., Hills, LV., Clague,-JJ., 1997. New acceleratormass spectrometry radiocarbon ages for the Mazama tephralayer from Kootenay National Park, British Columbia, Canada.Can. I Earth Sci. 34, 1202—1209.

Hansen, B.S., Engstrom, D.R., 1996. Vegetation history of Pleasant Island, Southeastern Alaska, since 13,000 yr BR Quat.Res. 46, 161—175.

Hebda, RI., 1995. British Columbia vegetation and climatehistory with focus on 6 ka BR Géogr. Phys. Quat. 49, 55—79.

Hebda, R.3., Allen, G.B., 1993. Modem pollen spectra from westtentral British Columbia. Can. J. Bot. 71, 1486—1495.

Hofmann, W., 1983. Stratigraphy of Cladocera and Chironomidae in a core from a shallow north German lake. Hydrobiologin 103, 235—239.

Holland, 5.5., 1976. Landforms of British Columbia: a physiographic outline. B.C. Dep. Mines Pet. Resour. Bull. 48, 138pp.

Levesque, A.3., Mayle, F.E., Walker, JR., Civynar, L.C., 1993.A previously unrecognized late-glacial cold event in easternNorth America. Nature 361, 623—626.

Lloyd, D., Angrove, K., Hope, 0., Thompson, C., 1990. A guideto site identification for the Kamloops Forest Region. B.C.

• Ministry of Forests, Victoria, B.C., 399 pp.Luckman, B.H., Holdsworth, G., Osborn, G.D., 1993. Neoglacial

glacier fluctuations in the Canadian Rockies. Quat. Res. 39,144—153.

MacIc RN., Rutter, N.W, Bryant Jr., V.M., Valastro, S., 1978a.Late Quaternary pollen record from Big Meadow, Pend Orielle

• County, Washington. Ecology 59, 956—965.Mack, R.N., Rutter, N.W., Valastro, S., 1978b. Late Quaternazy• pollen record from the Sanpoil River Valley, Washington. Can.

I. Bot. 56, 1642—1650.Mack, R.N., Rutter, N.W., Valastro, 5., 1979. Holocene vegeta

tion history of the Okanogan Valley, Washington. Quat. Res.12, 212—225.

Mann, D.H., Hamilton, T.D., 1995. Late Pleistocene mid

Holocene paleoenvironments of the North Pacific Coast. Quat.Sci. Rev. 14, 449—471.

Mathewes, R.W, 1973. A palynological study of postglacialvegetation changes in the University Research Forest, southwestern British Columbia. Can. I. Bot. 51, 2085—2103.

Mathewes, R.W., 1985. Paleobotanical evidence for climaticchange in southern British Columbia during late-glacial andHolocene time. In: Harington, C.R. (Ed.), Climate ChaiI~e inCanada, 5. Critical Periods in the Quaternary Climatic Historyof Northwestern North America. Syllogeus 55, 397—422.

Mathewes, R.W., Heusser, L.E., 1981. A 12,000 year palynological record of temperature and precipitation trends insouthwestern British Columbia. Can. 3. Bot. 59, 707—710.

Mathewes, R.W., King, M., 1989. Holocene vegetation, climate,and lake-level changes in the Interior Douglas-fir Biogeodimatic Zone, British Columbia. Can. I. Earth Sci. 26, 1811—1825.

McAndrews, 3., .Berti, A., Norris, G., 1973. Key to the Quaternary pollen and spores of the Great Lakes Region. Life Sci.Misc. PubI., R. Ont. Mus., Toronto, ON, 64 pp.

Mehringer, P.3. Jr., 1985. Late Quaterhary pollen records fromthe Interior Pacific Northwest and Northern Great Basin ofthe United States. In: Bryant, V.M., Holloway, R.G. (Eds.),Pollen Records of Late-Quaternaiy North American Sediments. American Association of Stratigraphic PalynologistsFoundation, Dallas, Tex., pp. 167—190.

Meidinger, D., Pojar, 3., 1991. Ecosystems of British Columbia.B.C. Ministry of Forests, Victoria, B.C., 330 pp.

Moore, P.D,, Webb, J.A., Collinson, M.E., 1991. Pollen Analysis(2nd ed.). Blackwell, Cambridge, 216 pp.

Oliver, D.R., Roussel, M.E., 1983. The Insects and Arachnidsof Canada, Part 11. The Genera of Larval Midges of Canada,Diptera: Chironomidae. Agric. Can. Publ. 1746, 263 pp.

Pellatt, M.G., 1996. Postglacial Changes in Vegetation and Climate Near Treeline in British Columbia. Ph.D. thesis, SimonFraser Univ., Burnaby, B.C., 177 pp.

Pellatt, M.G., Mathewes, R.W., 1994. Paleoecology and postglacial tree line fluctuations on the Queen Charlotte Islands,Canada. Ecoscience 1, 71—81.

Pellatt, MA)., Mathewes, R.W., 1997. Holocene tree line andclimate change on the Queen Charlotte Islands, Canada. Quat.Res. 48, 88—99.

Pellatt, M.G., Mathewes, R.W., Walker, I.R., 1997. Pollen analysis and ordination of lake sediment-surface samples fromcoastal British Columbia, Canada. Can. I. Bot. 75, 799—814.

Porter, S.C., Demon, G.H., 1967. Chronology of neoglaciation inthe North American Cordillera. Am. I. Sci. 265, 177—210.

Quickfall, G.S., 1987. PaludilIcation and climate on the QueenCharlotte Islands during the past 8000 years. M.Sc. Thesis,Simon Fraser Univ., 99 pp.

Ryder, J.M., Thomson, B., 1986. Neoglaciation in the southernCoast Mountains of British Columbia: chronology prior to thelate-Neoglacial maximum. Can. 3. Earth Sci. 23, 273—287.

Stager, J.C., Mayewski, PA., 1997. Abrupt early to midHolocene climatic transition at the Equator and the Poles.Science 276, 1834—1836.

Stuiver, M., Braziunas, T.F., Becker, B., Kromer, B., 1991. Cli-

138 MALi Pet/an et aL !Falaeogeography, Palaeoclirnatology, Palaeoecotogy 141 (1998) 123—138

matic, solar, oceanic and geomagnetic influences on Late-Glacial and Holocene atmospheric ‘4C/’2C change. Quat.Res. 35, 1—24.

Walker, J.R., 1987. Chironomidae (Diptera) in paleoecology.Quat. Sci. Rev. 6, 29—40.

Walker, Lit, 1988. Late-Quatemaiy Palaeoecology of Chirononiidae (Diptera: Insecta) from Lake Sediments in BritishColumbia. Ph.D. Thesis, Simon Fraser Univ., Burnaby, B.C.,204 pp.

Walker, JR., 1990. Modern assemblages of arctic and alpineChironomidae as analogues for late-glacial communities. Bydrobiologia 214, 223—227.

Walker, I.R., MacDonald, G.M., 1995. Distributions of Chirononiidae (Insecta: Diptera) and other freshwater midges withrespect to treeline, Northwest Territories, Canada. Arct. Alp.Res. 27. 258—263.

Walker, LR., Mathewes, R.W., 1987a. Chironomidae (Diptera)~and postglacial climate at Marion Lake, British Columbia,Canada. Quat. Res. 27, 89—102.

Walker, I.R., Mathewes, R.W., 1987b. Chironomids, lake trophicstatus, and climate. Quat. Res. 28, 431—437.

Walker, I.R., Mathewes, R.W., 1989a. Chironomidae (Diptera)remains in surficial lake sediments from the CanadianCordillera: analysis of the fauna across an altitudinal gradient. 3. Paleolimnol. 2, 61—80.

Walker, I.R., Mathewes, R.W., 1989b. Early postglacial chironomid succession in southwestern British Columbia, Canada, andits paleoenvironmental significance. J. Paleolimnol. 2, 1—14.

Walker, lit, Levesque, A.3., Cwynar, L.C., .Lotter, A.F.,1997. An expanded surface-water palaeotemperature inferencemodel for use with fossil midges from eastern Canada. 3.Paleolimnol. 18, 165—178.

Warner, E.G., 1984. Late Quaternary palaeoecology of EasternGraham Island, Queen Charlotte Islands, British Columbia.Ph.D. Thesis, Simon Fraser Univ., 190 pp.

Wiederholm, T., 1983. Chironomidae of the FIolarctic Region,Keys and Diagnoses, Part 1. Larvae. Entomok Scand. Suppl.19, 457 pp.