High-resolution palynology,climate change and human impact on a late Holocene peat bog on Haida Gwaii,British Columbia,Canada

Abstract

Haida Gwaii (formerly Queen Charlotte Islands) is an archipelago on the outer central coast of British Columbia. Drizzle Bog on Graham Island is a Sphagnum-dominated peatland selected for a multiproxy palaeoecological analysis using pollen, spores and other microfossils to see whether climatic changes such as the ‘Mediaeval Climate Anomaly’ (MCA), ‘Little Ice Age’ (LIA) and post-LIA warming could be detected in this hypermaritime region. A 90-cm-long Wardenaar peat core was sliced into 1-cm-thick subsamples and dated using ²¹⁰Pb back to ad 1892, and four accelerator mass spectrometry (AMS) radiocarbon dates back to 1800 cal. yr BP. Low pollen accumulation rates between ~ad 1600 and 1875 support cool growing seasons during the LIA. At 34 cm (~1875), total pollen and spore accumulation rates increased dramatically, coinciding with the end of the LIA based on evidence from glacier recession in the Coast Mountains of British Columbia. We suggest that the increased pollen production is likely a reflection of climatic warming and increased vegetation productivity following the LIA. Direct human impact is only apparent above 12 cm depth (~1958) correlated with road building across the southwestern edge of the wetland.

Keywords

climate change Haida Gwaii late Holocene ‘Little Ice Age’Pacific Northwest peatlands pollen accumulation rate

Introduction

Peatlands are important archives of ecosystem dynamics, particularly ombrotrophic bogs, where rapid peat accumulation and a lack of turbation provide an ideal record for high-resolution palaeoecological studies (Moore and Bellamy, 1974). Peatland formation is driven by the climate, geology and hydrology at a site in a way that leads to excess surface water, which can be altered by changes in climate, vegetation succession and/or human-induced change (Blackford, 2000). On shorter timescales, allogenic processes can play a significant role on mire development (Chambers and Charman, 2004). Extra-local and regional vegetation changes can be inferred from aerial pollen deposition, while local vegetation and hydrological changes are often preserved in the peat stratigraphy and recorded by local pollen and other microfossils. Peatland archives of pollen, spores, testate amoebae and other non-pollen palynomorphs (NPPs) thus provide a detailed record of climate and other environmental changes.

Trends in vegetation development are commonly demonstrated using the relative percentages of microfossil taxa. Another method for palaeoreconstruction is the application of pollen accumulation rates (PARs), sometimes referred to as pollen influx. PARs differ from relative percentage in measuring the abundance of each taxon independently of every other taxon (Faegri et al., 1989). A handful of studies have investigated the use of PARs, mostly of coniferous tree species, as a proxy for climate change (Bennett and Hicks, 2005; Hicks et al., 2004). From a physiological standpoint, pollen production of conifer trees is a possible proxy for summer temperatures as male cones are formed in late summer of the year preceding their release (Jalkanen et al., 2008). Conifer pollen productivity has been found to strongly correlate with July and August temperatures of the preceding year (Huusko and Hicks, 2009). The nature of a peat core record does not allow for the extraction of annual changes (Hicks et al., 2004); however, multiyear to decadal changes in climate, especially trends in summer temperature, should be apparent in the record. Few palynological studies investigate PARs of the last 200–300 years in fine resolution (Desprat et al., 2003; Kuoppamaa et al., 2009) at a site that has been subjected to little historical human impact.

The Drizzle Bog core spans most of the last two millennia, where the latter half is characterized by three climatic phases. The ‘Mediaeval Climate Anomaly’ (MCA) was a period of periodically warm, dry conditions of variable duration and extent that differ dramatically by region and for the purposes of this study is defined as spanning from ad 900 to 1450, similar to estimates by Bradley et al. (2003) and Herweijer et al. (2007). The MCA is followed by the ‘Little Ice Age’ (LIA), a period of episodic cool climate events with glacial advances spanning from approximately ad 1450 to 1875 (Fritz, 2003; Grove, 1988). Following the LIA is the well-documented increase in global temperatures (IPCC, 2007; Jones et al., 1999) coming out of the LIA from the late 19th century to the present. It was not known whether the climate changes of the last millennium are recorded on hypermaritime Haida Gwaii. The aim of this paper is to provide a high-resolution reconstruction of late Holocene peatland development from an ombrotrophic Sphagnum bog through analysis of pollen and other microfossils. Our hypothesis is that the MCA, LIA and post-LIA climate changes should be recorded in the microfossil record and/or in peat stratigraphy.

Methods

Study area

The study site is a raised ombrotrophic Sphagnum bog on Haida Gwaii (formerly Queen Charlotte Islands), British Columbia, Canada (Figure 1). Haida Gwaii is the most isolated island group in Canada, located on the western edge of the continental shelf off the British Columbia central coast, 50–150 km from the mainland. The Drizzle Bog coring site lies on northern Graham Island 10 km south of Masset (Figure 2) and about 300 m northeast of Highway 16 (53.92772′N, 132.10574′W; 62 m a.s.l.). The lowlands of northern Graham Island are composed of a patchwork of open peatland, bog woodland and coniferous forest lying within the submontane wet hypermaritime variant of the Coastal Western Hemlock Zone (CWHwh1; Green and Klinka, 1994). The forest surrounding the site is dominated by Sitka spruce, western hemlock and western red cedar (Picea sitchensis, Tsuga heterophylla and Thuja plicata, respectively) with occasional shore pine (Pinus contorta var. contorta) and yellow cedar (Callitropsis nootkatensis). Recent logging and gravel extraction have created disturbed patches (Figure 2), where successional vegetation with abundant red alder (Alnus rubra) and shrubs dominate. The nearest weather station, Sewall Masset Inlet, describes a temperate climate with a mean annual temperature of 8.2° and average yearly precipitation of 1509 mm (Environment Canada, 2011). Drizzle Bog falls within the shore pine–black crowberry–tough peat moss bog complex (Wb51), in the classification of MacKenzie and Moran (2004). Vegetation within this peatland complex is adapted to nutrient poor conditions and minerotrophic species are generally absent. The surface vegetation is dominated by Sphagnum mosses, with abundant ericaceous shrubs, sedges and other herbaceous plants, interspersed by stunted conifers. Tough peat moss (Sphagnum austinii) is a typical indicator species for the Wb51 bog complex and is one of the dominant hummock-forming peat mosses. A number of ericaceous shrubs often colonize the tops of raised Sphagnum hummocks. The most frequently encountered is black crowberry (Empetrum nigrum), but Rhododendron groenlandicum, Kalmia microphylla, Vaccinium uliginosum and Vaccinium oxycoccus are also present.

Figure 1.

Regional map showing the location of the Drizzle Bog study site and the location of studies discussed in text. (A) Bella Coola region (Desloges and Ryder, 1990), (B) Tide and Summit Lakes (Clague et al., 2004; Clague and Mathewes, 1996), (C) Wrangell Mountain region (Davi et al., 2003), (D) Western Prince William Sound and Gulf of Alaska (Calkin et al., 2001; Wiles et al., 1999) and (E) Queen Elizabeth Islands, including Ellesmere and Devon Islands (Bradley, 1990; Douglas et al., 1994; Fisher et al., 1983).

Figure 2.

Map of location and study site on northern Graham Island, Haida Gwaii.

Sampling and geochemical analysis

In the summer of 2003, a 90-cm peat core was retrieved by Rolf Mathewes and R.B. Worth using a Wardenaar corer (10 cm × 10 cm × 100 cm; Wardenaar, 1987) as part of a collaborative study. The total peat depth at the coring site is 2.83 m, but the upper 90 cm was selected to capture the period of interest. A total of 86 subsamples of 2 mL by displacement were taken at approximately 1 cm intervals for microfossil analysis. Preliminary physical and geochemical analysis of the Drizzle Bog core was conducted between March 2004 and February 2005 at the University of Heidelberg, Germany, under the direction of Dr William Shotyk (now at the University of Alberta), for the investigation of heavy metal deposition in peat. Only the peat bulk density and ash content percentage data were used in this study. The dry bulk density (g/cm³) was determined as per Givelet et al. (2004). Detailed geochemical analysis is being continued by W. Shotyk and will be published separately.

Microfossil analysis

Subsamples were prepared as per Faegri et al. (1989) using a modified standard preparation procedure that included 10% HCl to remove carbonates, 10% KOH to remove humic acids and acetolysis for the removal of cellulose. Lycopodium marker spores (11,300 ± 400, Batch no. 201890) were added to each subsample prior to processing in order to calculate pollen concentrations. Samples were mounted in silicone oil and counted at a magnification of 500× and at 1250× for critical identification. Pollen sums include all terrestrial pollen types, and all counts exceeded 450 grains (461–872). Microfossil identifications were made to the lowest taxonomic levels using the extensive reference collection at Simon Fraser University and a range of morphological keys (McAndrews et al. (1973), Moore et al. (1991) and Kapp et al. (2000) for pollen and spores; Ogden and Hedley (1980) and Payne et al. (2012) for testate amoebae and Van Geel (1976) for other NPPs). A number of ericaceous plants grow on or near the site, including at least three Vaccinium species (V. uliginosum, V. oxycoccus and V. caespitosum), Gaultheria shallon, R. groenlandicum and E. nigrum. Of these species, the Empetrum and Rhododendron pollen tetrads were identified separately due to size differences comparable to a study of the pollen morphology of Ericaceous plants from Burns Bog, British Columbia, by Hebda (1979). At the time the study was conducted, Empetrum was more abundant at Drizzle Bog, yet both E. nigrum and R. groenlandicum are associated with dry hummocks on coastal peat bogs, which justify splitting pollen identifications between Empetrum/Rhododendron and other undifferentiated Ericales pollen types for this study. Although pollen analysis was the primary goal of the study, testate amoebae, fungi and copepod spermatophores were analysed as supplementary evidence of environmental change.

The relative abundance of each taxon was calculated as a percentage of total terrestrial pollen. An abundance of each taxon was also calculated for each subsample as a PAR (Birks and Birks, 1980). The PAR is defined as the number of pollen grains per unit surface area per unit time (grains/cm²/yr). PARs were calculated for pollen and spores as well as other microfossils using the following equation: PAR = P_conc/D (where P_conc is pollen or microfossil concentration and D is the deposition time). The pollen concentration was determined using the exotic marker technique (Benninghoff, 1962), while the deposition rate was determined from the age–depth model. Stratigraphic diagrams were produced using TILIA v1.7 (Grimm, 2011), and pollen taxa were only included when their percentage abundance was >3% at any point in the core. Biostratigraphic pollen zones were determined with stratigraphically constrained clustering using the incremental sum of squares (CONISS; Grimm, 1987) after square root transformation of the pollen percentages (Edwards and Cavalli Sforza’s chord distance; Edwards and Cavalli-Sforza, 1964). Stratigraphic zones in the text are numbered and described in ascending order starting at the bottom of the core.

Chronology

The chronology for the peat core was based on 30 Lead-210 (²¹⁰Pb) ages and four accelerator mass spectrometry (AMS) radiocarbon (¹⁴C) dates. ²¹⁰Pb dates were determined for each sample down to 31.2 cm by Peter G. Appleby of the University of Liverpool as described in Appleby et al. (1997) using a constant rate of supply model (Appleby and Oldfield, 1978). Radiocarbon dates were obtained from peat samples with the uppermost date (44–46 cm) on Sphagnum by Beta Analytic Inc., Florida, USA (Table 1). Calibrated ages were calculated in Oxcal 4.0 (Bronk Ramsey, 2008) using the Northern Hemisphere terrestrial IntCal09 calibration curve (Reimer et al., 2009).

Table 1.

¹⁴C dates for the Drizzle Bog core. Calibrated ages of the 2σ confidence interval have been included and converted to yr ad and yr BP.

Material		Depth (cm)	Dates (¹⁴C yr BP)	Calibrated age
Material		Depth (cm)	Dates (¹⁴C yr BP)	2σ (95%) yr ad	2σ (95%) yr BP
Sphagnum moss	Beta 228502	44–46	180 ± 40	1650–1880	300–60
Peat	Beta 283271	62–64	1170 ± 40	770–980	1180–970
Peat	Beta 228504	80–82	1590 ± 40	390–560	1560–1390
Peat	Beta 283272	88.4–89.5	1820 ± 40	90–320	1860–1630

The age–depth relationship was calculated using a smoothed spline in the Clam package (Blaauw, 2010) within the statistical program R (R Development Core Team, 2011) to combine the ²¹⁰Pb age estimates with the ¹⁴C age estimates (Figure 3). Point estimates for each subsample were provided by the weighted mean of all age-model iterations (30,000) at each sample depth, and thus, all dates mentioned are given in calibrated years.

Figure 3.

Age–depth model of Drizzle Bog in cal. yr BP. Black line is the spline interpolation while the grey envelope shows the 95% confidence intervals. Black histograms represent calibrated curves for each of the four radiocarbon dates. Inset is the top 32 cm in calibrated years (ad) using 30 ²¹⁰Pb dates.

Results

Stratigraphy of the core reveals two clear units (Figure 4). The top half (0–45 cm) is a reddish-brown peat largely made up of well-preserved Sphagnum. The darker, more humified bottom half (45–90 cm) is composed of herbaceous material mixed with Sphagnum remains. This visual differentiation in the core is emphasized by a dramatic change in bulk density (Figure 4). The most obvious change occurs at 52 cm where bulk density changes from around 0.10 g/cm³ to just below 0.07 g/cm³ in the upper samples. Ash percentages (Figure 4) were only available for a portion of the core, but reveal a peak of 10% around 52 cm corresponding with the bulk density decline, much higher than the ~2% average for the rest of the core. Pollen concentrations (Figure 4) show a dramatic increase at this 52-cm mark to more than 125,000 grains/cm³ before dropping to concentrations between 10,000 and 25,000 grains/cm³.

Figure 4.

Peat stratigraphy of Drizzle Bog core with physical measurements and summary pollen abundances.

Pollen percentage diagram

The pollen percentage diagram was divided into three biostratigraphic zones (Figure 5) using CONISS. Zone DB 1 spans from 90 to 45.5 cm, while zone DB 2 extends up to 25.5 cm. Zone DB 3 spans from 25.5 cm to the modern surface and is divided into DB 3a and DB 3b subzones at 8.5 cm.

Figure 5.

Pollen percentage diagram for Drizzle Bog with pollen zones derived using CONISS. Stippling shows 5X exaggeration curves. The percentage of each taxon is based on the total terrestrial pollen sum for each sample. Sphagnum spores are not included in the pollen sum. Only those taxa with percentages greater than 3% in at least one sample are displayed.

Arboreal pollen percentages remain relatively constant throughout zone 1, while the percentage of Empetrum/Rhododendron pollen begins a gradual increase from <10% at 60 cm to >20% by the end of zone 1. Cyperaceae pollen increases from ~10% at the base to around 25%, where it remains relatively constant until a notable decline at 57 cm to <15% for the remainder of the core. Sphagnum spores frequently exceed 20% at the bottom of the core before an abrupt drop to their lowest percentage of <10% after 68 cm followed by a gradual increase to ~20% by the end of the zone.

The marked increase in Sphagnum spore percentage is the dominant feature of zone DB 2, as Sphagnum spores consistently make up over 40% compared with the terrestrial pollen percentage. Empetrum/Rhododendron pollen displays a notable peak of >35% immediately following 45 cm before gradually declining over the course of zone 2. Microseris borealis pollen peaks briefly at >25% before dropping to ~7% by the end of zone 2 possibly indicating the sample included flowering remains.

Pollen zone DB 3 is split into two subzones at 8.5 cm largely recognized by a doubling of Alnus pollen from a steady percentage of ~10% throughout the core to ~20% in subzone 3b. Empetrum/Rhododendron percentages have steadily declined after 45 cm to <5% through much of zone 3. Microseris borealis pollen shows an average of 7% in subzone 3a before a notable decline to <1% just prior to subzone 3b. Sphagnum spore percentages display an erratic series of peaks and valleys throughout subzone 3a before dropping to consistently remain at ~15% at the start of 3b for the duration of the core.

PAR diagram

The PAR diagram (Figure 6) differs dramatically from the percentage diagram. At the bottom of the core, the total PAR is below 1000 grains/cm²/yr and fluctuates between ~1000 and 2000 grains/cm²/yr until 55 cm. The total PAR then spikes at 52 cm, which is followed by low PARs of less than 1000 grains/cm²/yr from 50 to 37 cm. At this point, we observe a dramatic, fivefold increase to >6500 grains/cm²/yr peaking at 26 cm followed by a gradual decline to ~2000 grains cm²/yr at the top of the core.

Figure 6.

Pollen accumulation rate (grains/cm²/yr) diagram of Drizzle Bog with pollen zones and stippling showing 5× exaggeration curves. Pollen zones are those generated by the CONISS dendrogram on the pollen percentage diagram (Figure 5).

The proportional decline in Cyperaceae during DB 1 appears less dramatic in the PAR diagram when compared to the percentage diagram, yet Cyperaceae is still the most common pollen type early in the core compared with recent samples. This is apparent between 56 and 68 cm where a near absence of Sphagnum spores is displayed while Cyperaceae pollen is still relatively abundant. A spike in Tsuga heterophylla at 52 cm is noticeable, which was not immediately apparent from the percentage diagram, but the Empetrum/Rhododendron increase immediately following the transition to DB 2 is still evident. While Empetrum/Rhododendron pollen has a low accumulation rate at this time, many more pollen grains show up in relation to other taxa. Microseris borealis pollen is present in small numbers throughout the core but appears to be quite productive throughout most of DB 2 and DB 3a before declining dramatically in DB 3b. Gentiana douglasiana pollen mirrors M. borealis but remains present in DB 3b. The Alnus increase part way through DB 3b is evident in both diagrams suggesting a significant change in the regional vegetation.

NPP accumulation rate diagram

Two distinct periods of high fungal deposition are the major observable changes throughout zone DB 1. Increased deposition by Tilletia sphagni from the base of the core to 72 cm is partially concurrent with elevated deposition of type 18 spores from 83 to 52 cm. Copepod spermatophores are present in low numbers while testate amoebae are particularly infrequent below 55 cm.

Both fungal types along with copepod spermatophores show dramatic accumulation increases following periods of low abundance and peak near the end of zone DB 2. Most testate amoebae become more frequent at the same time, with Arcella artocrea notably averaging more than 50 tests/cm²/yr.

A defining feature of zone DB 3 is a decline in the accumulation of fungal types after ~20 cm. Nearly all testate amoebae increase to their highest accumulation rates coincident with the transition to subzone DB 3b. The exception is Arcella artocrea with a marked decline in this zone. The timing of the decline in A. artocrea matches the decline in Copepod spermatophores, with both of these types becoming infrequent near the top of the core.

Discussion

The development of mires along this temperate coast is a gradual progression from rich fen to poor fen to raised bog. A study by Quickfall (1987) investigated a peat exposure to the west of the Drizzle Bog core site and described this progression from the basal presence of rich fen species (Carex obnupta seeds and Lysichiton pollen) followed by increased concentrations of Ericales and Cyperaceae pollen, and finally by increasing Sphagnum spores. Another site describing similar late-stage mire development is Cottongrass Hollow on the Brooks Peninsula of Vancouver Island (Hebda, 1997). Despite initiating as a salt marsh, the progression from woody sedge peat to sedge peat to Sphagnum peat is documented. This mire development occurs as a gradient, and the later successional stages of poor fen and raised bog types share many of the same plant species. The bulk density of the mire (Figure 4) at the early portion of the core suggests a sedge-moss composition due to its higher bulk density (Chambers et al., 2011). Cyperaceae is also the most abundant pollen type, making up over 20% of all pollen until 57 cm, exceeding even Sphagnum spores for most of zone DB 1 with a sedge-moss composition.

The boundary between the two visually prominent peat types, along with the drop in bulk density after 52 cm is suggestive of a recurrence surface. A recurrence surface by definition is the boundary separating slow peat growth characterized by high humification, followed by fresh, actively growing Sphagnum peat (Birks and Birks, 1980). Immediately above 52 cm, we see a dramatic increase in peat accumulation. This high rate of peat growth suggests a transition to a raised bog with primarily ombrotrophic inputs. Sphagnum moss has become the dominant mire species as evidenced from the bulk density change (Figure 4) and the increase in Sphagnum spore abundance (Figures 5 and 6).

Two periods of dramatic depositional change of Sphagnum spores in the core are difficult to explain. The first period is a near absence of Sphagnum in zone DB 1 from 56 to 68 cm. This low abundance may be due to environmental conditions not conducive to Sphagnum spore production such as summer droughts prematurely drying out the spermatophytes (Sundberg, 2002). The second period of change is the erratic deposition of Sphagnum through subzone DB 3a (8.5–22.5 cm) that may be due to a turnover of Sphagnum species. The production and release of Sphagnum spores can vary by an order of magnitude between species (Gignac, 2001), which may partially explain the variations in Sphagnum spore abundance late in the core. Sphagnum spores were already abundant near the base of the core, and there is no subsequent loss of plant species in the pollen record. It seems reasonable that Drizzle Bog may already have been transitioning from a poor fen to a raised bog, with a distinct change to strictly ombrotrophic inputs after the recurrence surface at 52 cm.

The pollen concentration spike spanning 48–55 cm (Figure 4) is puzzling, and two scenarios are introduced as explanation. The first scenario is that the pollen concentration spike is linked to autogenous changes from a transition of the anoxic catotelm layer of the peat to the oxic acrotelm layer where aerobic decomposition takes over from anaerobic decomposition. The second scenario is that a prolonged period of dry conditions contributed to the pollen concentration spike. During prolonged dry periods peat dries out due to a lower than normal water-table, encouraging aerobic decomposition. If decomposition is equal to or greater than net primary production at the surface, peat growth can halt, functionally resulting in a hiatus in peat accumulation. As pollen is more resistant to decay than other vegetation components of peat, a large spike in pollen concentration likely indicates little to no peat accumulation or even peat decomposition. The spike in ash content percentage at 52 cm supports this interpretation since after organic material decomposes, the mineral components remain in the peat. As there are no relevant upland sources and no significant volcanic activity in the immediate area, it is likely that the ash content is derived nearly exclusively from the plant material. If the vegetation that normally contributes to peat accumulation decomposes, one would expect a high proportion of minerals compared to organic material, displayed by the ash content spike. This possibility of a hiatus puts the exact age of events between the top two radiocarbon dates (46 and 62 cm) into question. The uncertainty of ages for this part of the core means that the dating of any climate event linked to the pollen concentration spike is problematic. The most interesting changes in pollen deposition, however, occur in the latter half of the core after 46 cm and are thus unaffected by this possibility of a hiatus.

Human impact

Road construction

The only anthropogenic disturbance that appears to have impacted the mire community was the relatively recent construction of Highway 16 to the southwest of the coring site. The initial gravel road connecting Port Clements and Masset was built in ad 1958 (Leary, 1982), and paved in ad 1970, becoming the westernmost terminus of the Yellowhead Highway. The initial road construction slightly precedes pollen subzone DB 3b at 8.5 cm (ad 1967), and some noteworthy shifts in NPPs are apparent at this time. While the construction likely influenced bog hydrology and the local plant community, improved access to this formerly inaccessible area also influenced the regional pollen composition through logging and gravel extraction.

A single core is unlikely to be representative of the local hydrology as the surface wetness of a bog can vary considerably from place to place and over time. Despite this, the timing of the observed changes in NPP abundance does coincide with human impact that likely affected the bog’s hydrology. Copepods are an indicator of standing water (Anderson, 1998), and the decline of their spermatophores after this 12-cm mark may indicate a lowering of the water-table. The only identified species of testate amoeba that is reported to live in open water is Arcella artocrea (Payne et al., 2012) whose presence and subsequent decline mirrors the copepod spermatophore abundance at the same stage in the core. The high abundance of type 18 ascospores may also indicate the presence of standing water, as this fungal spore is commonly found in the presence of Eriophorum species (Van Geel, 1976). Both species of Eriophorum (E. chamissonis and E. angustifolium) found on Haida Gwaii can tolerate a wide range of wet substrates but are commonly found around the margins of ponds (Aiken et al., 2007). The large quantity of these three microfossil types together suggests wet conditions prior to the decline of all three to near absence following the road construction in ad 1958 (12 cm), though type 18 ascospores drop out of the record slightly earlier. There are currently ditches and culverts underneath the road, which allow the northeastern areas to drain resulting in the loss of pooling water. The increase in most other rhizopod taxa at this point may be a result of this reduced water-table as well. The lowering of the water-table would allow for more Sphagnum moss to colonize and grow, facilitating the increase in sphagnicolous rhizopods.

The most apparent feature defining pollen subzone DB 3b is a doubling in the percentage of Alnus pollen above 7 cm (Figure 5) corresponding to a date of ad 1972. The gravel road connecting Port Clements to Masset in ad 1958 opened this region, which was previously only reachable by boat, for clear-cut logging. Alder is one of the first trees to colonize following disturbance events that open up the forest canopy in this region. The alder pollen signal appears slightly delayed, which is to be expected, as the trees require time to colonize and mature.

Deer introduction

Sitka black-tailed deer (Odocoileus hemionus sitkensis) were introduced to Graham Island in the early 1900s (Dalzell, 1968), and are now the major large herbivore on the island. The decline in Empetrum/Rhododendron pollen and some of the herbaceous plants at the top of the core may be due to overgrazing (Gaston et al., 2008). Declines in ericaceous shrubs such as salal (G. shallon) and red huckleberry (V. parvifolium) as well as bog dwelling cloudberry (Rubus chamaemorus) have been specifically attributed to an overabundance of deer (Parks, 1992). Anecdotal observation suggests that deer are currently active around Drizzle Bog as fresh droppings were observed on a number of occasions.

Palaeoclimate interpretation

The PAR increase above 35 cm (ad 1870 ± 20) does not correspond with the major change in peat type (Figure 3), which occurs earlier at ~52 cm. No evidence of direct human impact prior to ad 1958 is apparent, suggesting that climate is likely the main driver influencing the changes in pollen accumulation.

LIA productivity

The majority of studies describing LIA climate throughout the Northern Hemisphere suggest periodic cool intervals between ad 1450 and 1900 (Bradley et al., 2003; Calkin et al., 2001; Fisher et al., 1983; Mann and Jones, 2003; Wiles et al., 1999). While global mean annual temperatures were reduced, ice core data from the Canadian high-arctic suggest that summer cooling may have been even greater than the drop in mean annual temperature (Fisher et al., 1983). The most recent period of low PARs in the Drizzle Bog core spans 48–35 cm with an age estimate of ad 1600–1875. Given the assumption that low total PAR is a proxy for cool summer temperatures, in this case below 1000 grains/cm²/yr, climatic conditions during the LIA appear to greatly reduce the regional pollen productivity of vegetation on and around Drizzle Bog.

Given the proposed influences of LIA climate on summer temperature, cold-adapted species are likely to have fared better over this period. This appears to be the case as the Empetrum/Rhododendron pollen type, largely made up of Empetrum nigrum pollen, shows a spike in pollen percentage spanning this time period. Empetrum nigrum is a dwarf shrub with a circumpolar distribution inhabiting a range of habitats from lowland Sphagnum bogs to subalpine regions on Haida Gwaii (Calder and Taylor, 1968). This wide tolerance range likely allows E. nigrum to cope with cool temperatures better than many of the other species that contribute to the regional pollen rain. Empetrum only shows a slight increase in PAR during the LIA, but displays a high proportional change in pollen productivity when compared with other taxa (Figure 5).

The timing of the LIA in the core compares reasonably well with other regional studies. Coast Mountain glaciers appear to have been at or near their maximum extent of the entire Holocene during this time (Desloges and Ryder, 1990), which the authors attributed to prolonged periods of below-average temperatures and/or well above average precipitation. Most of these Coast Mountain glaciers reached their maximum extent around ad 1875 followed by glacier retreat. Tree-ring accounts recording warm season temperatures (April–September) around the Gulf of Alaska record cool summers through much of this period with the coldest period occurring during ad 1790–1820 and subsequent warming around ad 1870 (Wiles et al., 1996). A tree-ring record of white spruce at the elevational treeline in the Wrangell Mountains (Davi et al., 2003) suggested two cold periods occurring around ad 1700 and between ad 1800 and 1830 with warming after ad 1850. This period of cool growing season conditions appears to have had a significant impact on regional pollen productivity. The low pollen accumulation between ~ad 1600 and 1875, nearly the lowest in the entire core, contrasts sharply with the pollen accumulation increase immediately following ~ad 1875.

Post-LIA productivity

Dramatic changes at Drizzle Bog that could be linked to modern climatic warming begin after ad 1875. The nearly fivefold increase in pollen productivity over a span of <30 years is noteworthy since this corresponds with other studies describing dramatic Northern Hemisphere warming after ~ad 1870 (Mann, 2002; Wiles et al., 1996; Wilson et al., 2006). Arctic communities have displayed similar responses to warming following the end of the LIA. Ice core data from the Agassiz and Devon ice caps from the Canadian high-arctic (Bradley, 1990) suggests that following a final cold period in ad 1820–1860, summer temperatures dramatically increased. This 19th-century warming is also evident from distinct changes to the diatom assemblage and increased productivity in adjacent sites of Ellesmere Island lakes (Douglas et al., 1994), which were attributed to longer growing seasons.

More regional examples of changes to biotic communities include lake cores from the northern Coast Mountains of British Columbia that show an increase in planktonic algae in the uppermost sediments (Clague and Mathewes, 1996). Increases in lake productivity inferred from high Pediastrum algal colonies were attributed to warming and subsequent lengthening of the growing season. Changes to the diatom assemblage in the 19th and 20th centuries in a follow-up study were suggestive of longer growing seasons and higher lake temperatures (Clague et al., 2004). Increases in copepod spermatophore abundance in the Drizzle Bog core (Figure 7) suggest a similar response. The extended length of the growing season likely boosted primary productivity of pools on the bog surface resulting in dramatic copepod increases, concurrent with increases in pollen productivity.

Figure 7.

Non-pollen palynomorph accumulation rate diagram of Drizzle Bog. Non-pollen palynomorphs other than testate amoebae are labelled as type identifiers used by Van Geel (1976). Accumulation rates are in grains/cm²/yr, while stippling shows 5× exaggeration curves. Pollen zones are those generated by the CONISS dendrogram on the pollen percentage diagram (Figure 5).

There are few palynological studies that investigate PARs over the last ~200 years in fine resolution. Barnekow et al. (2008) describe a change in pollen accumulation from lake sediments in northeast Sweden, and they attribute this increase either to land-use changes in the region or an over-estimation of PAR at the surface. While their speculation of land-use change is possible, the influence of climate also remains a possibility. Another Scandinavian study, from Northern Finland, describes a similar PAR increase after ad 1870 (Kuoppamaa et al., 2009). Prior to ~ad 1870, they attributed low PARs to colder climatic conditions corresponding to the LIA. Desprat et al. (2003) describe a nearly identical increase in pollen accumulation after ad 1860 from northeast Iberia. The authors suggest that recent warming and/or increasing CO₂ concentration most likely explains pollen production increases over the last ~150 years.

An important result from the Drizzle Bog core is that increases in pollen deposition are not limited to the regional arboreal taxa such as Pinus and Picea. Local bog taxa that are wind-disseminated (Empetrum, Cyperaceae and Sphagnum) and insect-pollinated (Gentiana douglasiana and Microseris borealis) display this rise in pollen productivity following ad 1875 (Figure 8). This suggests a change in climatic conditions, likely longer growing seasons and/or warmer summers, favourable to pollen and spore production in nearly all vegetation types. The effects of rising CO₂ concentrations, largely due to anthropogenic sources, is a possible factor since a number of studies describe fertilization effects of CO₂ on vegetation (Ainsworth and Long, 2005; Jablonski et al., 2002). However, while young trees reportedly grow faster and thus reach maturity quicker (Ladeau and Clark, 2006), there is little to no response of CO₂ enrichment on mature natural forests (Körner et al., 2005; Nowak et al., 2004; Figure 8).

Figure 8.

Pollen accumulation rates of taxa contributing to the region pollen grain as well as local bog taxa of the last ~400 years plotted on time and depth axes. Dashed line demarks the transition out of the LIA (~ad 1875) apparent in the core.

The recent increase in pollen accumulation after ~ad 1875 described from this isolated area suggests that climate warming is likely having a significant impact on regional pollen production. An indirect indicator of past pollen production is the timing of seasonal activities of plants. Trends in phenological change in Europe and North America clearly show a lengthening of the growing season (Menzel, 2000; Sparks and Menzel, 2002; Walther et al., 2002). Longer growing seasons result in increased primary productivity and a prolonged period of pollen production (Emberlin et al., 2002). Unfortunately, reliable phenological records generally start ad 1900 or later in western North America (Beaubien and Freeland, 2000), with scarce regional records for comparison.

Conclusion

The primary objectives of this study were to provide a reconstruction of the Drizzle Bog core to infer changes in regional vegetation and peatland development allowing for comparisons with other palaeoenvironmental reconstructions. The three factors that control peatland development, namely, vegetation succession, human impact and climate, all appear to have played a role.

The 90-cm core spanning ~1800 years records the successional stages of a hypermaritime peatland displaying transition from a slightly minerotrophic poor fen to an ombrotrophic raised bog. The autogenic driver of vegetation succession does not appear to play as strong a role as the allogenic factors of climate and anthropogenic impacts especially in the top 46 cm of the core. The inherent isolation of Haida Gwaii results in direct human impact not playing a role until very late in the core. Changes in peatland dynamics and microfossil assemblage due to road construction through Drizzle Bog are not apparent until after ad 1958. The regional impact due to logging accessibility in the area is also suggested from the recent doubling of the proportion of Alnus pollen, which is a clear indicator of disturbance.

Few climate-induced changes are evident before ad 1400 other than a period from ~ad 800 to 1200 where Sphagnum spore production appears to be nearly non-existent. A spike in pollen concentration at ~52 cm along with a peak in ash content suggests little to no peat growth, which may have been due to drier than normal conditions, but exact dating of the core during this period is problematic. Following this period, LIA influences are manifest as cool summers leading to low pollen productivity from ~ad 1600 to 1850, inferred by low accumulation rates. After ~ad 1875, PARs of all taxa increase dramatically, corresponding with the end of the LIA and subsequent warming. This apparent increase in pollen productivity in the majority of identified vegetation types is unprecedented in the 1800 years recorded in the core. Climate warming appears to be the primary driver of the increases described from this isolated region.

The results of this and other studies suggest the possibility that comparatively high concentrations of pollen may be a recent, post-LIA phenomenon. Due to the limited window of our knowledge of pollen productivity, biologically significant changes due to recent warming are likely to have begun earlier than previously thought.

Footnotes

Acknowledgements

The authors would like to thank the reviewers for their valuable comments that helped to considerably improve the quality of the manuscript. We thank R.B. Worth for his help in attaining the peat core. We are also grateful to Andriy Cheburkin for the ²¹⁰Pb measurements, Nicolas Givelet for sample preparation and density determinations, and Claudio Zaccone for ash contents.

Funding

This study was funded by the Natural Science and Engineering Research Council of Canada to R.W.M. (grant 3835).

References

Aiken

Dallwitz

Consaul

. (2007) Flora of the Canadian Arctic Archipelago: Descriptions, Illustrations, Identification, and Information Retrieval. Ottawa: NRC Research Press, National Research Council of Canada. Available at: http://nature.ca/aaflora/data (accessed 2 April 2012).

Ainsworth

Long

(2005) What have we learned from 15 years of free-air CO₂ enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO₂. New Phytologist 165: 351–371.

Anderson

(1998) A reconstruction of Holocene climatic changes from peat bogs in north-west Scotland. Boreas 27: 208–224.

Appleby

Oldfield

(1978) The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. Catena 5: 1–8.

Appleby

Shotyk

Fankhauser

(1997) Lead-210 age dating of three peat cores in the Jura Mountains, Switzerland. Water, Air, and Soil Pollution 100: 223–231.

Barnekow

Bragée

Hammarlund

. (2008) Boreal forest dynamics in north-eastern Sweden during the last 10,000 years based on pollen analysis. Vegetation History and Archaeobotany 17: 687–700.

BC Parks (1992) Naikoon Park master plan background document (draft). Prepared for BC Parks, Ministry of Environment, Lands and Parks, Canada, March, 82 pp.

Beaubien

Freeland

(2000) Spring phenology trends in Alberta: Links to ocean temperature. International Journal of Biometeorology 44: 53–59.

Bennett

Hicks

(2005) Numerical analysis of surface and fossil pollen spectra from northern Fennoscandia. Journal of Biogeography 32: 407–423.

10.

Benninghoff

(1962) Calculation of pollen and spores density. Pollen et Spores 4: 332–333.

11.

Birks

HJB

Birks

(1980) Quaternary Palaeoecology. London: Edward Arnold, 289 pp.

12.

Blaauw

(2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5: 512–518.

13.

Blackford

(2000) Palaeoclimatic records from peat bogs. Trends in Ecology & Evolution 15(5): 193–198.

14.

Bradley

(1990) Holocene paleoclimatology of the Queen Elizabeth Islands, Canadian High Arctic. Quaternary Science Reviews 9: 365–384.

15.

Bradley

Hughes

Diaz

(2003) Climate in medieval time. Science 302: 404–405.

16.

Bronk Ramsey

(2008) Deposition models for chronological records. Quaternary Science Reviews 27: 42–60.

17.

Calder

Taylor

(1968) Flora of the Queen Charlotte Islands: Part 1. Ottawa: Queen’s Printer and Controller of Stationary, 659 pp.

18.

Calkin

Wiles

Barclay

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

19.

Chambers

Charman

(2004) Holocene environmental change: Contributions from the peatland archive. The Holocene 14: 1–6.

20.

Chambers

Beilman

(2011) Methods for determining peat humification and for quantifying peat bulk density, organic matter and carbon content for palaeostudies of climate and peatland carbon dynamics. Mires & Peat 7: 1–10.

21.

Clague

Mathewes

(1996) Neoglaciation, glacier-dammed lakes, and vegetation change in northwestern British Columbia, Canada. Arctic Antarctic and Alpine Research 28(1): 10–24.

22.

Clague

Wohlfarth

Ayotte

. (2004) Late Holocene environmental change at treeline in the northern Coast Mountains, British Columbia, Canada. Quaternary Science Reviews 23(23–24): 2413–2431.

23.

Dalzell

(1968) The Queen Charlotte Islands 1774–1966. Terrace: C.M. Adam, 340 pp.

24.

Davi

Jacoby

Wiles

(2003) Boreal temperature variability inferred from maximum latewood density and tree-ring width data, Wrangell Mountain region, Alaska. Quaternary Research 60: 252–262.

25.

Desloges

Ryder

(1990) Neoglacial history of the Coast Mountains near Bella Coola, British Columbia. Canadian Journal of Earth Sciences 27: 281–290.

26.

Desprat

Sánchez Goñi

Loutre

M-F

(2003) Revealing climatic variability of the last three millennia in northwestern Iberia using pollen influx data. Earth and Planetary Science Letters 213: 63–78.

27.

Douglas

MSV

Smol

Blake

Jr (1994) Marked post-18th century environmental change in high-arctic ecosystems. Science 266: 416–419.

28.

Edwards

AWF

Cavalli-Sforza

(1964) Reconstruction of evolutionary trees. In: Heywood

McNeill

(eds) Phenetic and Phylogenetic Classification. London: The Systematics Association, pp. 67–76.

29.

Emberlin

Detandt

Gehrig

. (2002) Responses in the start of Betula (birch) pollen seasons to recent changes in spring temperatures across Europe. International Journal of Biometeorology 46: 159–170.

30.

Environment Canada (2011) Canadian climate normals. Available at: http://www.climate.weatheroffice.gc.ca/climate_normals/index_e.html

31.

Faegri

Kaland

Krzywinski

(1989) Textbook of Pollen Analysis. London: Alden Press, 328 pp.

32.

Fisher

Koerner

Paterson

WSB

. (1983) Effect of wind scouring on climatic records from ice-core oxygen-isotope profiles. Nature 301: 205–209.

33.

Fritz

(2003) Lacustrine perspectives on Holocene climate. In: Mackay

Battarbee

Birks

. (eds) Global Change in the Holocene. New York: Oxford University Press, pp. 227–241.

34.

Gaston

Golumbia

Martin

. (2008) Lessons from the Islands: Introduced species and what they tell us about how ecosystems work. Proceedings from the Research Group on Introduced Species 2002 Symposium, Queen Charlotte City, Canada, 1–5 October 2002. Ottawa: Canadian Wildlife Service, Environment Canada, 192 pp.

35.

Gignac

(2001) Bryophytes as indicators of climate change. The Bryologist 104(3): 410–420.

36.

Givelet

Le Roux

Cheburkin

. (2004) Suggested protocol for collecting, handling and preparing peat cores and peat samples for physical, chemical, mineralogical and isotopic analyses. Journal of Environmental Monitoring 6: 481–492.

37.

Green

Klinka

(1994) Land Management Handbook 28: A Field Guide for Site Identification and Interpretation for the Vancouver Forest Region. Victoria: Crown Publications, Inc., 285 pp.

38.

Grimm

(1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers & Geosciences 13: 13–35.

39.

Grimm

(2011) TILIA v1.7. Springfield, IL: Illinois State Museum, Research and Collections Centre.

40.

Grove

(1988) The Little Ice Age. New York: Methuen & Co, 498 pp.

41.

Hebda

(1979) Size, productivity, and paleoecological implications of ericaceous pollen from Burns Bog, southern Fraser River Delta, British Columbia. Canadian Journal of Botany 57: 1712–1717.

42.

Hebda

(1997) Late Quaternary paleoecology of Brooks Peninsula. In: Hebda

Haggarty

(eds) Brooks Peninsula: An Ice Age Refugium on Vancouver Island (Occasional Paper No. 5). Victoria: BC Parks, pp. 9.1–9.48.

43.

Herweijer

Seager

Cook

. (2007) North American droughts of the last millennium from a gridded network of tree-ring data. Journal of Climate 20: 1353–1376.

44.

Hicks

Goslar

van der Borg

(2004) A near annual record of recent tree-line dynamics from northern Finland. Acta Palaeobotanica 44(2): 299–316.

45.

Huusko

Hicks

(2009) Conifer pollen abundance provides a proxy for summer temperature: Evidence from the latitudinal forest limit in Finland. Journal of Quaternary Science 24(5): 522–528.

46.

IPCC (2007) Climate change 2007: Synthesis report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (Core Writing Team, eds Pachauri

Reisinger

), IPCC, Switzerland, 104 pp.

47.

Jablonski

Wang

Curtis

(2002) Plant reproduction under elevated CO₂ conditions: A meta-analysis of reports on 79 crop and wild species. New Phytologist 156: 9–26.

48.

Jalkanen

Hicks

Aalto

. (2008) Past pollen production reconstructed from needle production in Pinus sylvestris at the northern timberline: A tool for evaluating palaeoclimate reconstructions. Silva Fennica 42(4): 499–513.

49.

Jones

New

Parker

. (1999) Surface air temperature and its changes over the past 150 years. Reviews of Geophysics 37(2): 173–199.

50.

Kapp

Davis

King

(2000) Pollen and Spores. College Station, TX: The American Association of Stratigraphic Palynologists, 279 pp.

51.

Körner

Asshoff

Bignucolo

. (2005) Carbon flux and growth in mature deciduous forest trees exposed to elevated CO₂. Science 309: 1360–1362.

52.

Kuoppamaa

Goslar

Hicks

(2009) Pollen accumulation rates as a tool for detecting land-use changes in a sparsely settled boreal forest. Vegetation History and Archaeobotany 18: 205–217.

53.

Ladeau

Clark

(2006) Pollen production by Pinus taeda growing in elevated atmospheric CO₂. Functional Ecology 20: 541–547.

54.

Leary

(1982) More road building on Graham Island. In: Simpson

(ed.) Tales from the Queen Charlotte Islands Book 2. Masset: Senior Citizens of the Queen Charlotte Islands, pp. 55–58.

55.

McAndrews

Berti

Norris

(1973) Key to the Quaternary Pollen and Spores of the Great Lakes Region. Toronto: Royal Ontario Museum, 64 pp.

56.

MacKenzie

Moran

(2004) Wetlands of British Columbia: A Guide to Identification (Land Management Handbook). Victoria: Province of British Columbia, 287 pp.

57.

Mann

(2002) The value of multiple proxies. Science 297: 1481–1482.

58.

Mann

Jones

(2003) Global surface temperatures over the past two millennia. Geophysical Research Letters 30(15): 15–18.

59.

Menzel

(2000) Trends in phenological phases in Europe between 1951 and 1996. International Journal of Biometeorology 44: 76–81.

60.

Moore

Bellamy

(1974) Peatlands. London: Elek Science, 221 pp.

61.

Moore

Webb

Collinson

(1991) Pollen Analysis. Oxford: Blackwell Scientific Publications.

62.

Nowak

Ellsworth

Smith

(2004) Functional responses of plants to elevated atmospheric CO₂ – Do photosynthetic and productivity data from FACE experiments support early predictions? New Phytologist 162(2): 253–280.

63.

Ogden

Hedley

(1980) An Atlas of Freshwater Testate Amoebae. New York: Oxford University Press, 222 pp.

64.

Payne

Lamentowicz

van der Knaap

. (2012) Testate amoebae in pollen slides. Review of Palaeobotany and Palynology 173: 68–79.

65.

Quickfall

(1987) Paludification and climate on the Queen Charlotte Islands during the past 8000 years. MSc Thesis, Department of Biological Sciences, Simon Fraser University, Canada, 99 pp.

66.

R Development Core Team (2011) R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing.

67.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51(4): 1111–1150.

68.

Sparks

Menzel

(2002) Observed changes in seasons: An overview. International Journal of Climatology 22: 1715–1725.

69.

Sundberg

(2002) Sporophyte production and spore dispersal phenology in Sphagnum: The importance of summer moisture and patch characteristics. Canadian Journal of Botany 80: 543–556.

70.

Walther

G-R

Post

Convey

. (2002) Ecological responses to recent climate change. Nature 416: 389–395.

71.

Wardenaar

ECP

(1987) A new hand tool for cutting peat profiles. Canadian Journal of Botany 65: 1772–1773.

72.

Wiles

Barclay

Calkin

(1999) Tree-ring-dated Little Ice Age histories of maritime glaciers from western Prince William Sound, Alaska. The Holocene 2: 163–173.

73.

Wiles

D’Arrigo

Jacoby

(1996) Temperature changes along the Gulf of Alaska and the Pacific Northwest coast modeled from coastal tree rings. Canadian Journal of Forest Research 26: 474–481.

74.

Wilson

Wiles

D’Arrigo

. (2006) Cycles and shifts: 1,300 years of multi-decadal temperature variability in the Gulf of Alaska. Climate Dynamics 28(4): 425–440.

75.

Van Geel

(1976) A Paleoecological Study of Holocene Peat Bog Sections, Based on the Analysis of Pollen, Spores and Macro- and Microscopic Remains of Fungi, Algae, Cormophytes and Animals. Amsterdam: Hugo de Vries Laboratorium Universiteit, 75 pp.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB