Contrasting tundra responses to fertilization
5
6
7
2
3
4
8
9
10
11
12
1 Species compositional differences on different-aged glacial landscapes drive contrasting responses of tundra to nutrient addition
Sarah E. Hobbie 1* , Laura Gough 2 , and Gaius R. Shaver 3
1 Department of Ecology, Evolution and Behavior
University of Minnesota
1987 Upper Buford Circle
St. Paul MN 55108
2 University of Texas at Arlington, Biology Department
Box 19498
13
14
15
Arlington, TX 76019-0498
3 Ecosystems Center
21
22
23
18
19
20
16
17
Marine Biological Laboratory
7 MBL St.
Woods Hole MA 02543
* to whom correspondence should be addressed. email: shobbie@tc.umn.edu
FAX: 612-624-6777
1

Contrasting tundra responses to fertilization
1 Summary
2
3
4
1. In the northern foothills of the Brooks Range, Alaska, moist non-acidic tundra dominates more recently deglaciated upland landscapes, while moist acidic tundra dominates older upland landscapes. In previous studies, experimental fertilization of moist acidic tussock tundra greatly
5
6
7 increased the abundance and productivity of the deciduous dwarf shrub, Betula nana . However, this species is largely absent from moist non-acidic tundra.
2. These two common upland tundra community types exhibited markedly different responses to
8
9
10 fertilization with nitrogen and phosphorus. In moist acidic tundra, cover of deciduous shrubs
(primarily Betula nana ) increased after only two years, and by four years vascular biomass and aboveground net primary productivity (ANPP) had increased significantly, almost entirely
11
12 because of Betula . In moist non-acidic tundra, both biomass and ANPP were again significantly greater, but no single species dominated the response to fertilization. Instead, the effect was due
13
14
15 to a combination of several small, sometimes statistically non-significant responses by forbs, graminoids and prostrate deciduous shrubs.
3. The different growth form and species responses suggest that fertilization will cause carbon
16
17 cycling through plant biomass to diverge in these two tundra ecosystems. Already, production of new stems by apical growth has increased relative to leaf production in acidic tundra, while
18
19
20 the opposite has occurred in non-acidic tundra. Secondary stem growth has also increased as a component of primary production in acidic tundra, but is unchanged in non-acidic tundra. Thus, fertilization will likely increase carbon sequestration in woody biomass of Betula nana in acidic
21
22 tundra, while increasing carbon turnover (but not storage) of non-woody species in non-acidic tundra.
2

Contrasting tundra responses to fertilization
1 4. These results indicate that nutrient enrichment can have very different consequences for plant
2
3
4 communities that occur on different geologic substrates, because of differences in composition, even though they share the same regional species pool. Although the specific edaphic factors that maintain compositional differences in this case are unknown, variation in soil pH and related
5
6
7 variability in soil nutrient availability may well play a role.
Keywords
Alaska, Arctic, Betula nana , fertilization, moist acidic tundra, moist non-acidic tundra, net
8 primary production, nitrogen, pH, phosphorus
3

Contrasting tundra responses to fertilization
1 Introduction
2
3
4
Numerous studies have demonstrated that low nutrient availability strongly and consistently limits net primary production in tundra ecosystems (e.g., Henry et al., 1986;
Bowman et al., 1993; Chapin et al., 1995). In Alaskan upland tundra, woody deciduous shrubs,
5
6
7 particularly Betula nana , dominate the positive productivity response to increased nutrients
(Chapin & Shaver, 1985; Chapin et al., 1995; Bret-Harte et al., 2001; Shaver et al., 2001; Bret-
Harte et al., 2002). Warming manipulations, of both air and soil, also increase Betula nana
8
9
10 abundance and production in Alaskan tundra (Hobbie & Chapin, 1998, G. R. Shaver, unpublished data), likely, at least in part, because warming enhances nutrient availability.
Betula 's capacity to respond to nutrient addition is related to its growth habit of producing both
11
12 long and short shoots (Bret-Harte et al., 2001): Betula increases its production and biomass through conversion of short shoots into long shoots, followed by an increased rate of secondary
13
14
15 stem growth in this greater number of long shoots. Its many short shoots confer developmental plasticity that allows it to be more responsive to environmental change than other common tundra species (Bret-Harte et al., 2001).
16
17
Although Betula nana dominates the response to nutrient addition in studies of Alaskan upland tundra, Betula is not ubiquitous in upland tundra ecosystems in Alaska and the
18
19
20 circumpolar Arctic. On Alaska's North Slope, Betula is common in moist acidic tundra on mesic slopes of older landscapes that have become acidified through the processes of weathering and paludification (Walker et al., 1994; Walker et al., 1995). However, on circumneutral soils that
21
22
23 are rich in base cations, either because of recent glaciation or because of high inputs of loess,
Betula is quite rare, and moist non-acidic tundra, dominated by sedges and forbs, is common on mesic slopes (Walker et al., 1994; Walker et al., 1995). Recently glaciated landscapes are
4

Contrasting tundra responses to fertilization
1 common in the foothills region of the Brooks Range because of the expansion of mountain
2
3
4 glaciers during the Pleistocene (Hamilton, 2002) and loess-influenced circumneutral landscapes are common on Alaska's coastal plain (Walker et al., 1998; Walker & Everett, 1991). Moist acidic and moist non-acidic tundra are the two most common tundra types on Alaska’s North
5
6
7
Slope, making up 39% and 17%, respectively, of the vegetation in the Upper Kuparuk River
(Walker, 1998).
It is unclear how upland mesic communities without Betula nana will respond to nutrient
8
9
10 enrichment. In synthetic analyses of studies across the Arctic that included sites other than moist acidic tundra, fertilization with N alone or with P and K did not consistently cause deciduous shrubs to dominate (Dormann & Woodin, 2002), even in mesic sites (van Wijk et al., 2003). For
11
12 example, in northern Swedish tundra, responses by several species and growth forms or by graminoids were responsible for the effects of nutrient enrichment (Jonasson, 1992; Press et al.,
13
14
15
1998; Graglia et al., 2001; van Wijk et al., 2003), even in tundra types where Betula nana was common (Jonasson, 1992). In other Alaskan tundra types (e.g. wet sedge and heath tundra), increased biomass and production with fertilization again primarily resulted from a graminoid
16
17 response (Shaver et al., 1998; Gough et al., 2002).
Our objective was to compare the nutrient enrichment response of moist acidic and moist
18
19
20 non-acidic tundra in the foothills of the Brooks Range, Alaska. Because these two common tundra types can occur in close proximity, we were able to compare them between sites that were less than two km apart. Thus, the sites differed in parent material (because of different glacial
21
22
23 histories) but had similar climate and potential species pool, allowing us to overcome some of the limitations of syntheses of studies across broad regions where differences in many factors might influence community response to fertilization. Although the response to fertilization has
5

Contrasting tundra responses to fertilization
1 been assessed for these two tundra types separately (e.g. Shaver et al., 2001; Gough & Hobbie,
2
3
4
2003), they have never been directly compared using fertilization experiments conducted over the same time interval. We hypothesized that nutrient enrichment would increase plant biomass to a greater extent in acidic than in non-acidic tundra, because of the presence of Betula nana in
5
6
7 acidic tundra and the lack of overstorey shrubs in non-acidic tundra.
Methods
8
9
10
Fertilization with both nitrogen (N) and phosphorus (P) was initiated in two consecutive years (1996 and 1997) and repeated each June as part of the core Arctic Long-Term Ecological
Research (LTER) program. Sites are located on north-facing gentle slopes on the north and
11
12 south sides of Toolik Lake (68° 38'N, 149° 43'W, elevation 760 m), in moist non-acidic and acidic tundra, respectively. The moist non-acidic tundra site is located on the younger Itkillik II
13
14
15 glacial surface where the substrate is 11,500-25,000 years old, while the moist acidic site is located on the older Itkillik I glacial surface where the substrate is 50,000-120,000 years old
(Hamilton, 2002). Soils at the younger site are circumneutral, with relatively high concentrations
16
17 of exchangeable calcium, but slow rates of net N mineralization (Hobbie & Gough, 2002;
Hobbie et al., 2002). Plant biomass is generally less in non-acidic than acidic tundra (Walker et
21
22
23
18
19
20 al., 1995), but species diversity is higher, with more abundant and diverse forbs and graminoids
(Walker et al., 1994; Gough et al., 2000; Hobbie & Gough, 2004). Moist acidic tundra has soils with pH=3-4, but the faster rates of net N mineralization rates than in moist non-acidic tundra are reflected in higher foliar N concentrations (Hobbie & Gough, 2002; Hobbie et al., 2002).
Control and NP fertilized plots (5 m x 20 m) were established in replicate blocks (n=4 at the acidic site, n=3 at the non-acidic site). Fertilized plots annually received 10 g/m 2 N as
6

Contrasting tundra responses to fertilization
1 NH
4
NO
3
and 5 g/m 2 P as triple superphosphate in granular form in early June beginning in 1996
2
3
4 at the acidic site and in 1997 at the non-acidic site. In July of 1998-2001, aerial cover of the dominant growth forms was assessed using visual cover estimates in eight permanently located 1 m 2 quadrats in each plot. Each 1 m 2 quadrat was subdivided into 20 x 20 cm subquadrats in
8
9
10
5
6
7 which cover for each vascular species, mosses (as a group), and lichens (as a group) was estimated as well as for other ground categories (e.g. frost boil); these estimates were summed to achieve an absolute cover estimate for the 1 m 2 quadrat. Relative percent cover of each species was then determined by dividing the cover of the individual species by the total plant cover (sum of all vascular plant species, mosses and lichens) recorded for that 1 m 2 plot. These values represent canopy values only; mosses or other plants growing below taller plants were not
11
12 included. Our previous work has shown that this method of assessing cover provides similar relative composition of growth forms as biomass harvests, except that it underestimates woody
13
14
15 biomass of shrubs (Gough & Hobbie 2003). We use these cover estimates to provide a sense of how the canopy is changing in response to fertilization on a year-to-year basis, and over a relatively large portion of each plot, to supplement more quantitative but less frequent and
16
17 aerially representative biomass harvests. The same investigator supervised measurements in all years and standardized estimates across groups of data collectors. In 2000 and 2001, destructive
18
19
20 harvests were used to quantify the response to the treatments in terms of plant biomass and aboveground net primary production. Detailed harvest methods are contained in Shaver and
Chapin (1991). Briefly, in early August of each year, four 400 cm 2 quadrats containing plants
21
22
23 and soils were cut from the tundra to a depth below the rhizomes in each plot. Quadrats were taken back to the laboratory at the Toolik Field Station for sorting into live aboveground and belowground (rhizomes only, roots excluded) biomass by species. Bryophytes were sorted into
7

Contrasting tundra responses to fertilization
1 Sphagnum spp. and "other species" at the acidic site and Tomenthypnum nitens and "other
2
3
4 species" at the non-acidic site. Lichens were not sorted by species. In the case of deciduous shrubs, aboveground biomass was further sorted into current year's biomass and previous years' biomass on the basis of terminal bud scars and leaves. For sedges that produce overwintering
5
6
7 leaves (e.g. Eriophorum spp.), we classified all aboveground biomass as belonging to the current year, leading to slight overestimates of current year's biomass for these species. For shrubs, leaves were separated from stems, and for all species, reproductive parts (inflorescences, culms)
8
9
10 were removed from non-reproductive tissues. Live mosses were distinguished from dead mosses on the basis of colour and tissue integrity. Sorted tissues were dried at 65°C and weighed.
Botanical authorities can be found in Hultén (1968).
11
12
Aboveground vascular net primary production (ANPP) was estimated by summing current year's biomass, which includes both apical stem production and leaf production, as well
13
14
15 as a negligible amount of production invested in reproduction, for all species. In addition, aboveground secondary stem production (abbreviated SNPP below) was estimated using the equations of Bret-Harte et al. (2002) and included in estimates of total ANPP. Specifically, we
16
17 assumed that the percentages of old stem biomass that were secondary stem growth in the control and NP treatments respectively were 15.8 and 44.1 for Betula nana , 18.1 and 25.1 for Salix
18
19
20 pulchra , and 9.0 and 21.7 for Ledum palustre (Bret-Harte et al., 2002). We assumed that all other species of Salix encountered in our harvests were similar to Salix pulchra in their patterns of secondary stem growth, and that Vaccinium uliginosum and Rhododendron lapponicum
21
22 resembled Ledum palustre . In addition, we assumed that Vaccinium vitis-idaea , Cassiope tetragona , Dryas integrifolia , Arctostaphylos alpina , Andromeda polifolia , and Empetrum
8

Contrasting tundra responses to fertilization
1 nigrum produce negligible secondary stem growth (Shaver 1986), and we did not include them in
2
3
4 our estimates of secondary stem production.
Tissue nitrogen (N) concentrations were determined on biomass harvested in 2001 individually for several of the dominant species ( Eriophorum vaginatum , Carex bigelowii ,
5
6
7
Betula nana and Cassiope tetragona ) and for the remaining species pooled by growth form.
Separate analyses were done for current year’s leaves, current year’s stems, old leaves, old stems and rhizomes. Inflorescences were excluded from nutrient analyses, as their biomass is
8
9
10 negligible. All N analyses were conducted on a Costech ECS4010 Element Analyzer
(COSTECH Analytical, Valencia, California) at the University of Nebraska, Lincoln.
Subsets of the data presented here have been presented elsewhere. Biomass and ANPP
11
12 data from the non-acidic 2000 harvest were presented in comparison with additional treatments by Gough and Hobbie (2003). ANPP (averaged over the 2000 and 2001 harvests) from the
13
14
15 control plots were presented in Hobbie and Gough (2004). However, this is the first time that the two sites have been compared directly in terms of the response of biomass and productivity to fertilization using experiments conducted over the same time period.
16
17
Statistical Analyses
Relative cover was compared between sites and treatments and among dates using two-
18
19
20 way repeated-measures ANOVA on arcsine-square-root transformed data with year as a repeated measure. We compared biomass and production between sites, treatments and years using twoway repeated-measures ANOVA. Biomass was ln-transformed and quadrats were averaged
21
22
23 within plots before analyses. Tissue N concentrations for species and pooled growth forms and weighted average N concentrations and total aboveground N pools for all growth forms were compared between sites and treatments using two-way ANOVA. All statistical analyses were
9

Contrasting tundra responses to fertilization
1 conducted in JMP v. 5.0 (SAS Institute, Inc.). Repeated measures analyses were done fitting a
2
3
4
MANOVA model.
Results
Cover of all growth forms responded differently to NP fertilization at the two sites, as
5
6
7 indicated by significant interactions between site and treatment (Table 1). Deciduous shrub cover more than doubled with fertilization at the older, acidic site, but responded little to fertilization at the younger site (Fig. 1). In contrast, cover of graminoids and forbs increased at
8
9
10 the non-acidic site, but forbs declined and graminoids declined at least initially at the acidic site with fertilization. Cover of evergreen shrubs, mosses and lichens decreased with fertilization at both sites, but to differing degrees.
11
12
Total biomass and ANPP were greater at the older, acidic site (Tables 2, 3; Figs. 2, 3).
Total biomass did not change with fertilization at either site because of compensatory increases
13
14
15 in vascular plant biomass and decreases in moss and lichen biomass at both sites. Significant increases in total vascular biomass resulted from statistically non-significant increases in graminoid and forb biomass at the younger, non-acidic site, and in deciduous shrub biomass at
16
17 the older, acidic site (Fig. 2). Growth form responses to fertilization were a result, in turn, of statistically non-significant increases in biomass of several species with fertilization at the non-
18
19
20 acidic site (Table 2) and significantly greater biomass of Betula nana at the acidic site (Table 3).
ANPP increased markedly with fertilization at both sites in absolute magnitude and as a proportion of total biomass (Fig. 3, see Table S1 in Supplementary Material). However, the
21
22
23 significantly greater ANPP resulted from significant increases in forb (note site by treatment interaction) and deciduous shrub growth and statistically non-significant increases in graminoid growth at the non-acidic site, but increases in deciduous shrub growth at the acidic site.
10

Contrasting tundra responses to fertilization
1 Responses at the non-acidic site resulted from trends towards increased production by several
2
3
4 species in different growth forms, including graminoids (e.g. Carex spp.
, Eriophorum vaginatum ssp. triste ), deciduous shrubs (e.g. Dryas integrifolia ), evergreen shrubs (e.g. Rhododendron lapponicum ) and forbs (Table 2), whereas the significantly greater production of one species, the
5
6
7 deciduous shrub, Betula nana accounted for almost all the response at the acidic site (Table 3).
The lack of significant year by treatment or year by treatment by site interactions for biomass and ANPP suggests that our comparison of the treatment effects between sites is valid, despite
8
9
10 treatments being initiated one year apart.
The differences in growth forms responding to fertilization resulted in the principal components of total ANPP (leaf production, apical stem production, and secondary stem
11
12 production) reacting differently to fertilization at the two sites. Total apical ANPP (leaf production plus apical stem production) increased with fertilization at both sites (see Table S1).
13
14
15
However, at the non-acidic site, where herbaceous species and shrubs that produce little woody biomass (e.g. Dryas ) responded, leaf production dominated the apical production response to fertilization, as seen by an increase in the ratio of new leaf:new stem biomass (Fig. 4), in contrast
16
17 to a decrease at the non-acidic site, where apical stem production by Betula nana dominated the response (Fig. 4, see Table S1). In addition, secondary stem production as a component of total
18
19
20
ANPP (SNPP) increased with fertilization at the acidic site, because of the large increase in stem biomass associated with increased abundance of Betula nana at the acidic site, along with a high percentage of investment in secondary stem production by this species (Bret-Harte et al., 2002),
21
22
23 but remained unchanged at the non-acidic site.
Not surprising, fertilization with N and P significantly increased tissue N concentrations in nearly all species and growth forms (Tables 4, 5). For species and growth forms that exhibited
11

Contrasting tundra responses to fertilization
1 significant site differences in tissue concentrations, N was lower at the non-acidic site, consistent
2
3
4 with previous work at these sites (Hobbie & Gough, 2002). Total aboveground N pools of graminoids and deciduous shrubs increased with fertilization at both sites, even where the biomass of these growth forms did not respond to fertilization (Table 5). Total aboveground N
5
6
7 pools of forbs and mosses increased only at the non-acidic site, despite a significant decline in moss biomass at that site. In total, aboveground N pools were greater and increased more with fertilization at the acidic site.
8
9
10
Growth forms differed markedly in the ratio of ANPP to total aboveground and leaf N mass, with graminoids and forbs producing much more ANPP per mass of total and leaf N than evergreen shrubs, and deciduous shrubs being intermediate (Table 6). In addition, deciduous
11
12 shrubs had significantly greater ANPP per mass of total and leaf N at the non-acidic site compared to the acidic site, and both deciduous and evergreen shrubs had greater ANPP per
13
14
15 mass of leaf N at the non-acidic site. Fertilization reduced the ratio of ANPP to total aboveground and leaf N mass for all growth forms.
16
17
Discussion
Species and growth form compositional differences between moist acidic and non-acidic
18
19
20 tundra led to very different biomass, productivity and N responses by vascular plants to fertilization with N and P. As found in other studies of moist acidic tussock tundra at the Arctic
LTER site (Chapin et al., 1995; Shaver et al., 2001), fertilization led to a marked increase in the
21
22
23 cover, biomass and productivity of the woody shrub, Betula nana . This increase was restricted to acidic tundra where it was rapid, with greater deciduous shrub cover apparent after only two years of treatment and vascular aboveground biomass increased by 80% after four years. By
12

Contrasting tundra responses to fertilization
1 contrast, at the moist non-acidic site, biomass increased by only about 24% after five years of
2
3
4 fertilization. Presumably these differences in response arise because of inherent differences in the species present at these two tundra types, despite their sharing a common regional species pool. Betula nana , because of developmental plasticity associated with the production of short
5
6
7 and long shoots and a high proportion of secondary stem growth, accrues substantial biomass in response to nutrient enrichment (Bret-Harte et al., 2001; Bret-Harte et al., 2002). By contrast, the vascular plant species that responded positively to fertilization at the non-acidic site were
8
9
10 either herbaceous graminoids and forbs, or prostrate deciduous and evergreen shrubs that produce little woody biomass (Gough & Hobbie, 2003). These results indicate that the moist non-acidic site may exhibit a response to fertilization more typical of other tundra sites, where it
11
12 is often found that several, rather than a single, species respond positively to fertilization, or that graminoids dominate the positive response (Graglia et al., 2001; Dormann & Woodin, 2002;
13
14
15
Gough et al., 2002; van Wijk et al., 2003).
The sites were similar in exhibiting declines in moss and lichen biomass with fertilization. Decreased biomass of mosses and lichens is often observed with fertilization (e.g.
16
17
Chapin et al. 1995; Press et al. 1998; Cornelissen et al. 2001) and such declines have been attributed to shading and/or burial by litter produced by greater vascular plant biomass. It is also
18
19
20 possible that the high levels of fertilizer applied directly to the moss and lichen surface was toxic to those plants.
The factors keeping the abundance of Betula nana low in moist non-acidic tundra are
21
22
23 presently unknown. The biogeochemistry of moist non-acidic tundra differs from acidic tundra in a number ways, with moist non-acidic tundra having neutral soils, lower N supply rates (but higher dissolved N pool sizes), higher P and Ca availability, lower rates of dissolved organic N
13

Contrasting tundra responses to fertilization
1 production and slower rates of soil and litter decomposition (Hobbie & Gough, 2002; Hobbie et
2
3
4 al., 2002; Hobbie & Gough, 2004; Nordin et al., 2004). In an associated study, L. Gough
(unpublished) found that Betula seeds were capable of germinating in non-acidic tundra when sown, with significantly greater germination rates in plots from which mosses and vascular
5
6
7 plants had been removed. These results suggest that existing non-acidic vegetation may be hampering germination of naturally wind-dispersed seeds and thus limiting colonization by
Betula .
8
9
10
In addition, even if capable of germinating, Betula nana may experience competitive exclusion, nutrient deficiencies or toxicity in non-acidic tundra where rates of N supply (both inorganic and organic) are much lower (Hobbie & Gough, 2002; Hobbie et al. 2002). The
11
12 deciduous shrubs in moist acidic tundra (which include Betula nana ), have lower productivity per unit mass of total aboveground or leaf N than the graminoids at either site and the deciduous
13
14
15 shrubs at the non-acidic site (Table 6). Thus, Betula may be less competitive at the non-acidic site where N availability is relatively low, although it might become more competitive following
N addition, if it were able to establish. Other factors besides low N availability may keep Betula
16
17 nana rare in non-acidic tundra, particularly if it possesses traits associated with calcifuge species generally (Lee 1999). For example, it may be susceptible to iron deficiency in circumneutral
18
19
20 soils. It may also lack Ca-sequestering mechanisms that enable it to avoid metabolically inhibitory concentrations of Ca in its cells when growing with relatively high Ca supply.
Although low P availability has been implicated in excluding calcifuge species from calcareous
21
22
23 soils in other studies (Lee 1999), it likely does not contribute to Betula ’s low abundance in nonacidic tundra, as extractable P concentrations are comparable or higher in non-acidic than in acidic tundra, especially in the mineral soil (Hobbie & Gough 2002).
14

Contrasting tundra responses to fertilization
2
3
4
1 The different growth form and species biomass responses to nutrient enrichment at the two sites imply that the dynamics of carbon cycling and storage through plants will diverge with continued nutrient addition, since plant biomass contains 40-60% carbon (data not shown). Stem biomass at the acidic site was more than double that at the non-acidic site under control
5
6
7 conditions, and nearly doubled in response to nutrient addition, while stem biomass showed little change with nutrient addition at the non-acidic site. In addition, the allocation of new production to stems increased relative to leaves with fertilization at the acidic site, suggesting that fertilized
8
9
10 plots will continue on a trajectory of increasing woody biomass and carbon storage in aboveground biomass (Mack et al., 2004). In contrast, at the non-acidic site, fertilization will likely lead to substantially less aboveground biomass accrual than it will at the acidic site, but
11
12 should increase the turnover of biomass (i.e. fertilization has increased the ratio of production:biomass). Whether these site differences in plant biomass accrual will translate into
13
14
15 differences in ecosystem carbon sequestration with fertilization is unclear—after 20 years of continuous fertilization, increased Betula nana biomass with fertilization at a nearby acidic tundra site was more than offset by more rapid organic matter decomposition in the soil (Mack et
16
17 al., 2004), despite the slow decay rate of woody stems. The mechanism behind this faster decomposition is unclear, but may result from direct stimulation of decomposition by high
18
19
20 nutrient availability (Hobbie & Gough, 2004) and faster decomposition of more N-rich tissues.
Decomposition rates at the non-acidic tundra are inherently slower than they are in acidic tundra, but also respond positively to nutrient addition (Hobbie & Gough, 2004).
21
22
23
The differences in response to nutrient enrichment also have implications for how fertilization will affect the energy balance of these different tundra ecosystems. An increase in biomass and shrub density decreases the albedo of tundra, increasing sensible heat flux.
15

Contrasting tundra responses to fertilization
1 However, it decreases ground heat flux, because of shading of the soil surface by the shrub
2
3
4 canopy, which in turn reduces the active layer depth (McFadden et al., 1998; Chapin et al., 2000;
Thompson et al., 2004). Thus, fertilization may lead to a more marked change in the soil thermal environment in acidic tundra, where it causes a large change in vegetation structure, than in non-
5
6
7 acidic tundra. In addition, increased shrub density increases snow accumulation in winter, warming winter soils, potentially increasing winter decomposition rates (Walker et al., 1999;
Sturm et al., 2001). Fertilization of non-acidic tundra, in contrast, should not alter winter snow
8
9
10 accumulation.
Because future climate warming is expected to increase soil N availability in arctic tundra
(Nadelhoffer et al., 1992; Hobbie et al., 2002), the results presented here offer insights into how
11
12 these communities may respond to climate warming. Although the amount of fertilizer added annually in this study undoubtedly exceeds the increase in nutrient availability that will occur
13
14
15 with future climate warming, the response to fertilization could indicate the direction, if not the magnitude, of vegetation change that will occur with such warming. The interpretation that climate warming might therefore be expected to cause divergent responses in these two common
16
17 upland tundra vegetation types (increasing deciduous shrub abundance in moist-acidic tundra, while increasing graminoid and forb abundance in moist non-acidic tundra) is partly supported
18
19
20 by results from warming manipulations conducted at these two sites as well as by paleoecological data. Warming increased Betula nana abundance at the acidic site after three years (Hobbie & Chapin, 1998). However, four years of warming had little effect on community
21
22
23 composition at the non-acidic site (Gough & Hobbie, 2003), though greater vole herbivory in the experimental greenhouses used to increase temperature in that study clouded the interpretation of their effects on vegetation dynamics (Gough & Hobbie, 2003). Also, during early Holocene
16

Contrasting tundra responses to fertilization
1 warming, vegetation on older glacial surfaces experienced an increase in Betula abundance,
2
3
4 while non-acidic tundra vegetation composition of the younger, Itkillik II glacial surface has remained relatively stable since deglaciation (Oswald et al., 2003).
5
6
7
Acknowledgments
We are grateful to all of the “pluckers” who helped with the biomass harvests and plant cover censuses including: J. Hawkins, M. Hedlund, K. Howell, L. McSherry, T. Miley, M. Moss, N.
8
9
10
Pauliukonis, S. Pratt, J. Schimtz, M. van der Welle, P. Vermeulen, and M. Weiss. We also thank the Toolik Lake Field Research Station. This research was supported by a collaborative grant from the National Science Foundation (
OPP-9902695 to S. E. Hobbie and OPP-9902721 to L.
11
12
13
14
Gough) and by
the Arctic LTER (DEB-9810222).
Supplementary Material
The following material is available from
15
16
17 http://www.blackwellpublishing.com/products/journals/suppmat/JEC/JEC-CODE/JEC-
CODE.htm
18
19
Table S1. Components of plant biomass and production in control and fertilized (NP) treatments
(2000-2001).
17

Contrasting tundra responses to fertilization
1 References
2
3
4
Bowman, W.D., Theodose, T.A., Schardt, J.C., & Conant, R.T. (1993) Constraints of nutrient availability on primary production in two alpine tundra communities. Ecology, 74, 2085-
8
9
10
11
12
5
6
7
2097.
Bret-Harte, M.S., Shaver, G.R., & Chapin, F.S., III (2002) Primary and secondary stem growth in arctic shrubs: implications for community response to environmental change. Journal of Ecology, 90, 251-267.
Bret-Harte, M.S., Shaver, G.R., Zoerner, J.P., Johnstone, J.F., Wagner, J.L., Chavez, A.S.,
Gunkelman, R.F., IV, Lippert, S.C., & Laundre, J.A. (2001) Developmental plasticity allows Betula nana to dominate tundra subjected to an altered environment. Ecology, 82,
18-32.
13
14
15
Chapin, F.S., III, Eugster, W., McFadden, J.P., Lynch, A.H., & Walker, D.A. (2000) Summer differences among arctic ecosystems in regional climate forcing. Journal of Climate, 13,
2002-2010.
16
17
Chapin, F.S., III & Shaver, G.R. (1985) Individualistic growth response of tundra plant species to environmental manipulations in the field. Ecology, 66, 564-576.
18
19
20
Chapin, F.S., III, Shaver, G.R., Giblin, A.E., Nadelhoffer, K.J., & Laundre, J.A. (1995)
Responses of arctic tundra to experimental and observed changes in climate. Ecology, 76,
694-711.
21
22
23
Cornelissen, J. H. C., Callaghan, T. V., Alatalo, J. M., Michelsen, A., Graglia, E., Hartley, A. E.,
Hik, D. S., Hobbie, S. E., Press, M. C., Robinson, C. H., Henry, G. H. R., Shaver, G. R.,
Phoenix, G. K., Gwynn Jones, D., Jonasson, S., Chapin, F. S., III, Molau, U. Lee, J. A.,
18

Contrasting tundra responses to fertilization
1
2
3
4
Melillo, J. M., Sveinbjornsson, B., & Aerts, R. (2001) Global change and arctic ecosystems: is lichen decline a function of increases in vascular plant biomass? Journal of
Ecology, 89, 984-994.
Dormann, C.F. & Woodin, S.J. (2002) Climate change in the arctic: using plant functional types
5
6
7 in a meta-analysis of field experiments. Functional Ecology, 16, 4-17.
Gough, L. & Hobbie, S.E. (2003) Responses of moist non-acidic arctic tundra to altered environment: Productivity, biomass and species richness. Oikos, 103, 204-216.
8
9
10
Gough, L., Shaver, G.R., Carroll, J., Royer, D.L., & Laundre, J.A. (2000) Vascular plant species richness in Alaskan arctic tundra: the importance of soil pH. Journal of Ecology, 88, 54-
66.
11
12
Gough, L., Wookey, P.A., & Shaver, G.R. (2002) Dry heath arctic tundra responses to long-term nutrient and light manipulations. Arctic, Antarctic and Alpine Research, 34, 211-218.
13
14
15
Graglia, E., Jonasson, S., Michelsen, A., Schmidt, I.K., Havstrøm, M., & Gustavsson, L. (2001)
Effects of environmental perturbations on abundance of subarctic plants after three, seven and ten years of treatments. Ecography, 24, 5-12.
16
17
Hamilton, T.D. (2002). Glacial geology of the Toolik Lake and Upper Kuparuk River region. In
(ed D.A. Walker). University of Alaska Fairbanks, Insitute of Arctic Biology, Fairbanks,
18
19
20
AK.
Henry, G.H.R., Freedman, R., & Svoboda, J. (1986) Effects of fertilization on three tundra plant communities of a polar desert oasis. Canadian Journal of Botany, 64, 2502-2507.
21
22
Hobbie, S.E. & Chapin, F.S., III (1998) The response of tundra plant biomass, aboveground production, nitrogen, and CO
2
flux to temperature manipulation. Ecology, 79, 1526-1544.
19

Contrasting tundra responses to fertilization
1 Hobbie, S.E. & Gough, L. (2002) Foliar and soil nutrients in tundra on glacial landscapes of
2
3
4 contrasting ages in northern Alaska. Oecologia, 131, 453-463.
Hobbie, S.E. & Gough, L. (2004) Litter decomposition in moist acidic and non-acidic tundra with different glacial histories. Oecologia, 140, 113-124.
5
6
7
Hobbie, S.E., Miley, T.A., & Weiss, M.S. (2002) Carbon and nitrogen cycling in soils from different glacial surfaces in northern Alaska
. Ecosystems, 5, 761-774.
Hultén, E. (1968) Flora of Alaska and neighboring territories Stanford University Press,
8
9
10
Stanford, California, USA.
Jonasson, S. (1992) Plant responses to fertilization and species removal in tundra related to community structure and clonality. Oikos, 63, 420-429.
11
12
Lee, J.A. (1999) The calcicole-calcifuge problem revisited. Advances in Botanical Research, 29,
1-30.
13
14
15
Mack, M.C., Schuur, E.A.G., Bret-Harte, M.S., Shaver, G.R., & Chapin, F.S., III (2004)
Ecosystem carbon storage in arctic tundra reduced by long-term fertilization. Nature, 431,
440-443.
16
17
McFadden, J.P., Chapin, F.S., III, & Hollinger, D.Y. (1998) Sub-grid scale variability in the surface energy balance of arctic tundra. Journal of Geophysical Research-Atmospheres,
18
19
20
21
22
103, 28947-28961.
McKane, R.B., Johnson, L.C., Shaver, G.R., Nadelhoffer, K.J., Rastetter, E.B., Fry, B., Giblin,
A.E., Kielland, K., Kwiatkowski, B.L., Laundre, J.A., & Murray, G. (2002) Resourcebased niches provide a basis for plant species diversity and dominance in arctic tundra.
Nature, 415, 68-71.
20

Contrasting tundra responses to fertilization
5
6
7
2
3
4
1 Nadelhoffer, K.J., Giblin, A.E., Shaver, G.R., & Linkins, A.E. (1992). Microbial processes and plant nutrient availability in arctic soils. In Arctic ecosystems in a changing climate: an ecophysiological perspective (eds F.S. Chapin, III, R.L. Jefferies, J.F. Reynolds, G.R.
Shaver & J. Svoboda), pp. 281-300. Academic Press, San Diego, California, USA.
Nordin, A., Schmidt, I.K., & Shaver, G.R. (2004) Nitrogen uptake by arctic soil microbes and plants in relation to soil nitrogen supply. Ecology, 85, 955-962.
Oswald, W.W., Brubaker, L.B., Hu, F.S., & Kling, G.W. (2003) Holocene pollen records from
8
9
10
11
12 the central Arctic Foothills, northern Alaska: testing the role of substrate in the response of tundra to climate change. Journal of Ecology, 91, 1034-1048.
Press, M.C., Potter, J.A., Burke, M.J.W., Callaghan, T.V., & Lee, J.A. (1998) Responses of a subarctic dwarf shrub heath community to simulated environmental change. Journal of
Ecology, 86, 315-327.
13
14
15
Shaver, G. R. (1986) Woody stem production in Alaskan tundra shrubs. Ecology 67:660-669.
Shaver, G.R., Bret-Harte, M.S., Jones, M.H., Johnstone, J., Gough, L., Laundre, J., & Chapin,
F.S., III (2001) Species composition interacts with fertilizer to control long-term change
16
17 in tundra productivity. Ecology, 82, 3162-3181.
Shaver, G.R. & Chapin, F.S., III (1991) Production:biomass relationships and element cycling in
18
19
20
21
22 contrasting arctic vegetation types. Ecological Monographs, 61, 1-31.
Shaver, G.R., Johnson, L.C., Cades, D.H., Murray, G., Laundre, J.A., Rastetter, E.B.,
Nadelhoffer, K.J., & Giblin, A.E. (1998) Biomass accumulation and CO
2
flux in wet sedge tundras: responses to nutrients, temperature, and light. Ecological Monographs, 68,
75-97.
21

Contrasting tundra responses to fertilization
1 Sturm, M., Racine, C., & Tape, K. (2001) Climate change: Increasing shrub abundance in the
2
3
4
Arctic. Nature, 411, 546-547.
Thompson, C., Beringer, J., Chapin, F.S., III, & McGuire, A.D. (2004) Structual complexity and land-surface energy exchange along a gradient from arctic tundra to boreal forest. Journal
5
6
7 of Vegetation Science, 15, 397-406. van Wijk, M.T., Clemmensen, K.E., Shaver, G.R., Williams, M., Callaghan, T.V., Chapin, F.S.,
III, Cornelisen, J.H.C., Gough, L., Hobbie, S.E., Jonasson, S., Lee, J.A., Michelsen, A.,
8
9
10
Press, M.C., Richardson, S.J., & Rueth, H. (2003) Long-term ecosystem level experiments at Toolik Lake, Alaska, and at Abisko, Northern Sweden: generalizations and differences in ecosystem and plant type responses to global change. Global Change
11
12
Biology, 10, 105-123.
Walker, D.A. (1998)
The Geobotanical Atlas of the Toolik Lake and Kuparuk River Region,
13
14
15
16
17
Alaska . http://www.geobotany.uaf.edu/arcticgeobot/index.html
.
Walker, D.A., Auerbach, N.A., Bockheim, J.G., Chapin, F.S., III, Eugster, W., King, J.Y.,
McFadden, J.P., Michaelson, G.J., Nelson, F.E., Oechel, W.C., Ping, C.L., Reeburg,
W.S., Regli, S., Shiklomanov, N.I., & Vourlitis, G.L. (1998) Energy and trace-gas fluxes across a soil pH boundary in the Arctic. Nature, 394, 469-472.
18
19
20
21
Walker, D.A., Auerbach, N.A., & Shippert, M.M. (1995) NDVI, biomass, and landscape evolution of glaciated terrain in northern Alaska. Polar Record, 31, 169-178.
Walker, D.A. & Everett, K.R. (1991) Loess ecosystems of northern Alaska: Regional gradient and toposequence at Prudhoe Bay. Ecological Monographs, 61, 437-464.
22

Contrasting tundra responses to fertilization
1 Walker, M.D., Walker, D.A., & Auerbach, N.A. (1994) Plant communities of a tussock tundra
2
3
4 landscape in the Brooks Range Foothills, Alaska. Journal of Vegetation Science, 5, 843-
866.
Walker, M.D., Walker, D.A., Welker, J.M., Arft, A.M., Bardsley, T., Brooks, P.D., Fahnestock,
8
9
10
5
6
7
11
J.T., Jones, M.H., Losleben, M., Parson, A.N., Seastedt, T.R., and Turner, P.L. (1999)
Long-term experimental manipulation of winter snow regime and summer temperature in arctic and alpine tundra, Hydrologic Processes, 13, 2315-2330.
Received 13 September 2004; revision accepted 17 January, 2005
Handling Editor: Richard Bardgett
23

Contrasting tundra responses to fertilization
Table 1. P-values from Repeated-Measures ANOVAs analyzing relative percent cover data of growth forms for 1998, 1999, 2000, and 2001 in control and fertilized plots in acidic and non-acidic tundra for each of six growth forms. Data were arcsine-square root transformed for analyses. P-values <0.10 are indicated in bold.
Growth Form Site (S) S x T
Source
Year (Y) Y x S Y x T Y x S x T
Graminoid
Deciduous Shrub
Evergreen Shrub
Forb
Moss
Lichen
<0.001
<0.001
<0.001
<0.001
0.002
0.01
Treatment
(T)
0.14
<0.001
<0.001
0.51
<0.001
<0.001
0.005
<0.001
0.02
0.002
<0.001
0.009
0.002
0.01
<0.001
0.15
<0.001
0.04
0.25
0.10
0.004
0.15
0.50
0.58
0.03
0.07
0.19
0.21
0.42
0.07
0.02
0.009
0.001
0.46
0.16
0.36
24

Contrasting tundra responses to fertilization
Table 2. Current year's (new) biomass and total biomass of species at the non-acidic site measured in 2000 and 2001 in control and fertilized plots
(NP = nitrogen, phosphorus). New biomass includes leaves and aboveground stems produced during the current year. Total biomass includes total aboveground biomass and belowground stem biomass, but excludes roots. Statistical significance is presented for Repeated-Measures ANOVAs comparing current year's biomass or total biomass between treatments and years (with year as a repeated measure). Data were ln-transformed for analysis. Current year's biomass was not determined for mosses, as indicated by dashes.
New Biomass (g/m 2 ) Repeated Measures Significance 4 Total Biomass (g/m 2 ) Repeated Measures Significance 4
Species
Eriophorum vaginatum
2000
69.4±
11.8
63.5±
48.9
2001
71.2±
9.0
37.2±
28.7
CT NP CT NP Treatment
(T) ns
Source
Year
(Y) ns
Carex bigelowii
8.0±
2.8
73.2± 9.0±
67.3 1.6
57.1±
36.6 ns ns
2000 2001
T x Y CT NP CT NP Treatment
(T) ns ns
111.0
87.8± 126.8
47.6± ns
±
18.4
62.9 ±
12.4
36.5
24.8± 113.2
37.4± 113.3
ns
7.9 ±
102.6
5.8 ±
55.0
Source
Year
(Y) ns ns
T x Y ns ns
25

Contrasting tundra responses to fertilization
Arctogrostis ssp. triste
Carex
0.0± 2.9± 0.0± 4.9± latifolia
Eriophorum angustifolium
0.0 2.9 0.0 4.1
10.6± 20.5± 11.2± 33.6±
3.9 2.2 5.9 10.4
0.0± 8.8± 0.4± 7.0± membranacea 0.0 4.8 0.2 4.0
Carex vaginata 0.0± 6.3± 0.5± 8.6±
0.0 3.9 0.5 4.3
Carex spp.
1 0.3± 0.0± 1.0± 0.5±
0.2 0.0 0.5 0.3
Salix reticulata 2.4± 5.3± 2.3± 8.9±
1.1 4.0 0.2 4.2
Vaccinium 0.1± 3.3± 0.7± 1.1± uligonosum 0.1 2.4 0.7 0.9 ns ns ns ns ns ns ns ns ns
* ns ns ns ns ns ns
0.0± 5.1± 0.0± 7.2± ns
0.0
4.5
5.1
5.1
0.0
16.6± 33.1± 27.2± 55.9± ns
12.4
6.0
14.4
† ns ns ns ns
0.0± 13.0± 0.7± 13.3± ns
0.0 7.7 0.4 7.2
0.0± 8.6± 0.7± 13.8± ns
0.0 5.2 0.7 7.0
1.1±
0.8
0.0±
0.0
2.3±
0.8
0.7±
0.4
*
14.9± 18.5± 15.3± 40.6± ns
7.0 9.5 0.3 19.7
1.2± 13.5± 19.5± 7.0± ns
0.8 8.6 18.4 6.2 ns ns ns ns ns ns ns ns ns ns ns ns ns ns
26

Contrasting tundra responses to fertilization
Dryas 15.4± 31.5± 13.6± 25.1± integrifolia alpina
4.1
0.9
4.3
3.3
3.9
1.0
14.3
Arctostaphylos 1.7± 5.9± 2.0± 2.8±
2.4 vitis-idaea
Cassiope tetragona
0.0 0.0 0.0 0.0
6.3± 15.0± 7.6± 3.8±
2.5 5.9 1.9 1.2 ns ns
Salix arctica 2.8± 1.4± 2.4± 0.0±
2.8 1.2 1.2 0.0 ns
Ledum palustre 0.0± 0.0± 0.0± 0.0± —
0.0 0.0 0.0 0.0
Vaccinium 0.0± 0.0± 0.0± 0.0± — ns ns ns ns
†
Rhododendron 0.1± 0.8± 0.0± 1.1± *** lapponicum 0.1 0.0 0.0 0.2 ns
—
— ns ns ns
77.5± 76.1± 44.0± 44.2± ns
18.3 10.3 13.2 20.6
12.9± 17.5± 12.9± 13.9± ns
7.3 12.0 2.8 13.4
40.4± 21.1± 18.3± 0.0± ns
40.4 14.4 11.0 0.0
— 0.0± 0.0± 2.3± 1.2± —
0.0 0.0 1.5 1.2
— 0.3± 0.0± 0.0± 0.0± —
* ns
0.3 0.0 0.0 0.0
109.9
141.6
96.9± 53.6± ns
± ±
47.4 56.9
36.2 15.9
0.4± 3.9± 0.6± 4.2± **
0.4 0.9 0.5 0.4 ns
—
—
† ns ns ns
—
—
* ns ns ns ns
27

Contrasting tundra responses to fertilization
Andromeda polifolia
Polygonum bistorta
0.9±
0.9
0.3
0.0±
0.0
0.3± 1.2±
0.7
0.1±
0.1
0.0±
0.0
0.1± 1.4±
0.1 1.3 ns ns ns ns ns ns
4.2± 0.0± 0.4± 0.0± ns
4.2 0.0 0.4 0.0
1.0± 3.6± 0.5± 2.7± ns
0.8 2.7 0.5 2.5
6.2± 11.6± 6.4± 8.2± ns
3.7 7.5 3.4 5.1 ns
*
Equisetum spp. 2.8± 6.7± 3.6± 2.9±
1.6 4.5 1.8 1.5
Other forbs 2 2.2± 11.3± 3.2± 15.7±
Tomenthypnum
1.4 6.0 1.0 5.5
— — — — nitens
Other mosses 3 — — — — ns
†
—
— ns
*
—
— ns ns 4.9± 17.7± 7.1± 21.2± †
3.2 7.7 3.0 5.9
— 43.0± 42.8± 35.2± 10.4± ns
25.2 20.0 11.0 4.9
— 121.8
56.0± 61.4± 63.1± † ns
†
† ns
±
12.2
9.5 3.6 21.0
1 Unidentifiable species of Carex .
2 Other forbs include species of Cardamine , Claytonia , Lagotis , Pedicularis , Pyrola , Saussuria , Saxifraga , Thalictrum , and Tofieldia , and
Polygonum viviparum.
3 Other mosses include Hylocomium splendens , Aulacomnium spp., Dicranum spp.
28 ns ns ns ns
† ns

Contrasting tundra responses to fertilization
4 Statistical significance indicated as follows: ***P≤0.001, **P≤0.01, *P≤0.05, †P≤0.10, ns=P>0.10.
29

Contrasting tundra responses to fertilization
Table 3. Current year's (new) biomass and total biomass of species at the acidic site measured in 2000 and 2001 in control and fertilized plots (NP = nitrogen, phosphorus). New biomass includes leaves and aboveground stems produced during the current year. Total biomass includes total aboveground biomass and belowground stem biomass, but excludes roots. Statistical significance is presented for Repeated-Measures ANOVAs comparing current year's biomass or total biomass between treatments and years (with year as a repeated measure). Data were ln-transformed for analysis. Current year's biomass was not determined for mosses, as indicated by dashes.
New Biomass (g/m 2 ) Repeated Measures' Significance 4 Total Biomass (g/m 2 ) Repeated Measures' Significance 2
Species
2000 2001
CT NP CT NP Treatment
(T)
Source
Year
(Y) ns Eriophorum vaginatum
78.2±
22.8
71.5±
25.1
55.0±
5.2
51.9±
15.4 ns
Carex bigelowii
0.1±
0.1
Betula nana 25.7±
2.8
4.3±
3.5
78.4±
23.4
1.0±
0.6
13.3±
3.0
5.7±
2.7
56.5±
8.3
†
**
Salix pulchra 5.9±
Vaccinium uliginosum
4.3
0.7±
0.4
Rubus chamaemorus
6.4±
3.8
0.2±
0.2
3.9±
2.8
29.8±
18.0
0.5±
0.5
0.0±
0.0
6.3±
3.0
1.8±
1.8
0.4±
0.4
24.0±
11.9 ns ns ns
†
† ns
† ns ns ns ns ns ns
2000 2001
T x Y CT NP CT NP Treatment
(T) ns 120.4
±
35.2
2.3±
2.3
152.2
±
22.2
26.5±
24.1
3.4±
2.4
46.1±
30.0
94.4±
33.0
87.8±
12.6
9.0±
4.2
437.3
±
146.5
0.4±
0.4
13.6±
12.4
63.4±
37.0
4.4±
1.7
201.2
±
55.8
1.6±
1.6
0.0±
0.0
40.3±
23.4
81.8±
20.1 ns
16.3±
5.9
328.6
±
22.2
16.0±
16.0
3.5±
3.5
65.8±
29.3
†
* ns ns ns
Source
Year
(Y) ns
† ns ns
† ns
T x Y ns ns ns ns ns ns
30

Contrasting tundra responses to fertilization
Ledum palustre
Vaccinium vitis-idaea
12.4±
4.2
14.6±
4.4
12.7±
3.3
12.3±
3.9
7.9±
1.8
8.8±
1.1
12.3±
7.2
7.1±
2.2 ns ns ns ns ns ns
158.9
±
18.8
105.4
±
144.9
±
33.5
79.4±
27.1
180.2
±
24.9
86.7±
13.1
138.6
±
61.0
67.4±
20.5 ns ns ns ns
Cassiope tetragona
Empetrum nigrum
Andromeda polifolia
2.1±
0.7
0.6±
0.3
2.2±
1.1
3.4±
1.4
3.4±
2.0
0.6±
0.4
2.4±
0.9
0.4±
0.2
1.2±
1.2
2.5±
0.4
1.9±
1.2
1.7±
1.0
Polygonum bistorta
Pedicularis
0.3±
0.2
0.8±
0.0±
0.0
0.1±
0.2±
0.1
0.3±
0.1±
0.1
0.1± spp . 0.2 0.1 0.2 0.0
Sphagnum spp . — — — — ns ns ns ns
†
— ns ns ns ns
*
— ns ns ns ns
*
14.0
29.7±
11.8
5.4±
3.0
11.4±
5.8
1.0±
0.6
1.8±
0.5
— 68.9±
30.8±
11.5
29.7±
13.4
3.0±
1.8
0.0±
0.0
0.2±
0.2
6.3±
37.9±
14.1
4.0±
2.4
11.4±
11.4
0.3±
0.2
0.7±
0.5
19.3±
35.9±
7.1
17.8±
10.5
10.1±
5.8
0.1±
0.1
0.1±
0.1
1.2± ns ns ns ns
*
* ns ns ns ns
*
**
Other mosses 1 — — — — — — —
19.1
70.4±
1.8
31.5±
10.2
89.8±
0.4
13.8± ** †
9.4 4.4 15.7 5.9
1 Other mosses include Hylocomium splendens , Aulacomnium turgidum , Dicranum spp., Polytrichum spp., Ptilium crista-castrensis , Pleurozium ns ns ns ns ns ns
† ns
* shreberi , Tomenthypnum nitens .
2 Statistical significance indicated as follows: ***P≤0.001, **P≤0.01, *P≤0.05, †P≤0.10, ns=P>0.10.
31

Contrasting tundra responses to fertilization
Table 4. Biomass N concentrations (%) of species growing in control and fertilized plots in moist acidic and non-acidic tundra analyzed from the 2001 harvest. Common species of graminoids, and deciduous and evergreen shrubs were analyzed individually, while less common vascular species and mosses and lichens were pooled for analyses. Values are means ± SE. Significance of twoway ANOVAs with site (S) and treatment (T) as main effects is also presented. For species that occurred only at one site, treatments were compared using t -tests.
Species or group Tissue category
Leaf control
Acidic
NP
Non-acidic
Control NP
1.64 ± 0.04 2.32 ± 0.05 1.51 ± 0.02 2.15 ± 0.29
S
ANOVA Significance
T SxT
*** Eriophorum vaginatum
Carex bigelowii
Other graminoids
Rhizome
Leaf
Rhizome
Leaf
Rhizome
2.08 ± 0.07 4.01 ± 0.38 1.37 ± 0.18
2.04 ± 0.22 2.74 ± 0.23 2.41 ± 0.06
0.67 ± 0.05 1.22 ± 0.05 0.76 ± 0.04
--
--
--
--
1.87 ± 0.04
0.87 ± 0.08
2.16 ± 0.77
2.43 ± 0.09
1.06 ± 0.12
2.32 ± 0.23
1.49 ± 0.21
**
--
--
**
†
**
*
†
--
--
32

Contrasting tundra responses to fertilization
Betula nana Leaf
New stem
Old stem
Rhizome
Other deciduous shrubs
Leaf
Cassiope tetragona
New stem
Old stem
Rhizome
New leaf + stem
Old leaf + stem
Rhizome
Other evergreen shrubs New leaf
New stem
2.35 ± 0.10 3.75 ± 0.12
1.88 ± 0.13 2.44 ± 0.06
0.71 ± 0.04 1.08 ± 0.05
0.60 ± 0.03 1.08 ± 0.11
--
--
--
--
--
--
--
--
2.47 ± 0.40 3.39 ± 0.07 1.71 ± 0.05 2.41 ± 0.11
1.22 ± 0.06 2.46 ± 0.29 1.09 ± 0.05
0.91 1.15 ± 0.15 0.59 ± 0.04 1.15 ± 0.28
0.86 ± 0.07 1.51 ± 0.15 0.63 ± 0.01 1.08 ± 0.18
1.70 ± 0.25 3.30 ± 0.36 1.43 ± 0.08
0.79 ± 0.04 1.63 ± 0.19 0.75 ± 0.02
0.72 ± 0.01 0.86 ± 0.07 0.61 ± 0.01
1.59 ± 0.14 2.72 ± 0.11 1.49 ± 0.32
1.36 ± 0.08 2.06 ± 0.13 1.85 ± 0.36
1.75 ± 0.32
2.65 ± 0.19
1.49 ± 0.2
0.87 ± 0.11
3.00 ± 0.21
2.04 ± 0.11
--
--
--
--
**
*
**
***
***
***
**
***
**
**
***
**
***
*
--
--
--
--
33

Contrasting tundra responses to fertilization
Forbs
Mosses
Lichens
Old leaf
Old stem
Rhizome
Shoots
Rhizomes
1.14 ± 0.04 2.04 ± 0.19 0.88 ± 0.11
1.65 ± 0.19 2.77
2.22 ± 0.30
0.67 ± 0.03 1.10 ± 0.07 0.88 ± 0.17 1.15 ± 0.10
0.64 ± 0.01 0.94 ± 0.04 0.65 ± 0.09 0.84 ± 0.05
2.71 ± 0.33 5.02 1.65 ± 0.26 2.60 ± 0.43
0.96 ± 0.07 2.14 ± 0.57
0.87 ± 0.04 2.34 ± 0.19 0.84 ± 0.01 2.02 ± 0.08
1.15 ± 0.42 1.61 ± 0.06 0.95 ± 0.31 1.32 ± 0.12
Statistical significance indicated as follows: ***P≤0.001, **P≤0.01, *P≤0.05, †P≤0.10, ns=P>0.10.
For acidic NP, there was only one block with sufficient forb mass for tissue N analysis. Thus, no SE is presented.
--
--
***
**
***
--
--
***
--
--
34

Contrasting tundra responses to fertilization
Table 5. N concentration and total N mass for growth forms and total biomass sampled in 2001. Total biomass includes rhizomes but not roots. Aboveground biomass excludes rhizomes. N mass was determined by multiplying the N concentrations of individual tissues within species by their corresponding biomass and summing the products across species and growth forms. N concentration is a weighted average, determined by dividing by the N mass by the corresponding biomass value. Values are means ± SE. Significance of two-way ANOVAs with site (S) and treatment (T) as main effects is also presented.
S
ANOVA significance
%N N mass
T SxT S T SxT Growth form
Control
Acidic
NP
N (%) N
Mass
(g/m 2 )
N
(%)
N
Mass
(g/m 2 )
N (%)
Non-Acidic
Control NP
N
Mass
(g/m 2 )
N
(%)
N
Mass
(g/m 2 )
Graminoid 1.77 ± 1.57 ± 2.73 2.72 ±
0.05 0.26 ±
0.17
0.61
Deciduous shrub
0.80 ±
0.05
1.92 ±
0.60
1.57 6.58 ±
± 1.07
1.34 ±
0.06
0.81 ±
0.02
2.54 ±
0.29
± 1.77 4.18 ±
±
0.12
0.70
0.89 ± 1.47 1.49 ±
0.28 ± 0.15
*** ***
***
* *
**
*
** *
35

Contrasting tundra responses to fertilization
Evergreen shrub
0.12
0.78 ± 2.50 ± 1.39 3.66 ±
0.00 0.24 ±
0.07
0.90
0.19
0.76 ±
0.01
0.76 ± 1.26 0.70 ±
0.29 ±
0.15
0.12
Forb
Lichen
Moss
2.16 ± 0.02 ± 3.85 0.01 1.31 ±
0.30 0.01 0.16
0.18 ± 2.43 0.71 ±
0.02 ±
0.48
0.05
1.19 ±
0.41
0.63 ±
0.14 ±
1.61 0.24 ±
± 0.11
0.06
0.87 ± 0.94 ± 2.34 0.36 ±
0.04 0.15 ± 0.14
0.95 ±
0.31
0.35 ± 1.32 0.34 ±
0.10 ± 0.16
0.84 ±
0.01
0.81 ±
0.08
0.12
2.02 1.45 ±
± 0.30
0.19
Total aboveground
1.14 ± 3.11 ± 1.91 8.12 ±
0.02 0.06 ± 0.90 vascular 0.12
1.32 ±
0.06
0.08
2.46 ± 2.11 4.63 ±
0.11 ±
0.18
0.41
--
***
-- --
***
***
**
***
*
**
*** ***
***
**
*
36

Contrasting tundra responses to fertilization biomass
Total biomass
Total vascular biomass
0.92 ± 7.57 ± 1.69 13.57
0.03 0.47 ±
0.06
± 1.49
0.93 ± 6.00 ± 1.67 12.97
0.03 0.22 ±
0.06
± 1.45
1.00 ±
0.04
5.54 ± 1.67 8.87 ±
0.21 ±
0.12
0.87
1.06 ±
0.04
4.37 ±
0.25
1.65 7.08 ±
±
0.16
0.43
***
***
Statistical significance indicated as follows: ***P≤0.001, **P≤0.01, *P≤0.05, †P≤0.10, ns=P>0.10.
For acidic NP, there was only one block with sufficient forb mass for tissue N analysis. Thus, no SE is presented.
** ***
** *** *
37

Contrasting tundra responses to fertilization
Table 6. The ratio of apical ANPP to total aboveground vascular N mass and leaf N mass for growth forms and total biomass sampled in 2001. For these calculations, ANPP excludes reproductive tissues. Values are means ± SE. Significance of two-way ANOVAs with site (S) and treatment (T) as main effects is also presented.
Acidic Non-acidic
Control NP Control NP
ANPP/N mass (g g -1 y -1 )
Graminoid 60.79 ± 42.20 60.44 ± 46.70
ANOVA significance
S T SxT
***
0.21 1.50 ±
0.79
Deciduous shrub 25.81 ± 18.96
0.92 ±
1.42
Evergreen shrub 13.90 ±
1.72
9.72
±
Total aboveground 30.05 ± 20.06 vascular biomass
ANPP/leaf N (g g -1 y -1 )
1.80 ±
0.42
54.56 ±
1.97
14.57 ±
1.87
1.80
Forb 38.08 ± 19.90 64.33 ±
4.73 11.99
49.15 ±
4.48
±
4.48
42.04
±
4.92
10.52
±
1.02
40.37
±
5.89
41.69
±
3.17
***
--
***
**
*
--
**
--
38

Contrasting tundra responses to fertilization
Graminoid 60.79 ± 42.20
1.50 ±
0.79
Deciduous shrub 49.11 ± 34.80
60.44 ±
0.21
2.07 ±
0.81
Evergreen shrub 26.50 ± 17.17
76.22 ±
1.60
68.61 ±
2.38 ±
1.90
2.37
Forb 60.79 ± 19.90 64.33 ±
4.73 11.99
46.70
±
4.48
58.76
±
3.67
32.43
±
3.69
40.37
±
5.89
***
***
†
***
***
†
*** ***
Total aboveground 45.49 ± 31.96 vascular biomass 1.37 ±
63.03 ±
0.95
±
47.73
±
*** ***
1.08 4.12
Statistical significance indicated as follows: ***P≤0.001, **P≤0.01, *P≤0.05, †P≤0.10, ns=P>0.10.
For acidic NP, there was only one block with sufficient forb mass for tissue N analysis. Thus, no
SE is presented.
39

Contrasting tundra responses to fertilization
Figure Legends
Fig. 1. Relative cover of growth forms in control and N+P fertilized plots in moist non-acidic and moist acidic tundra over for years, from 1998-2001. Fertilization was initiated in 1996 in acidic tundra and in 1997 in non-acidic tundra.
Fig. 2. Total biomass (excluding roots) of growth forms in control and N+P fertilized plots in moist non-acidic and moist acidic tundra in 2000 and 2001. Statistical significance is indicated for a two-way repeated measures ANOVA. S (site), Y (year), T (treatment); † P≤0.1, * P≤0.05,
** P≤0.01, *** P≤0.001.
Fig. 3. Vascular plant aboveground net primary production (including estimated secondary stem production [wood]) of growth forms in control and N+P fertilized plots in moist non-acidic and moist acidic tundra in 2000 and 2001. Statistical significance is indicated for a two-way repeated measures ANOVA. S (site), Y (year), T (treatment); † P≤0.1, * P≤0.05, ** P≤0.01, ***
P≤0.001.
Fig. 4. The ratio of new leaf biomass to new stem biomass in in control and N+P fertilized plots in moist non-acidic and moist acidic tundra in 2000 and 2001.
40

Contrasting tundra responses to fertilization
Figure 1
41

Contrasting tundra responses to fertilization
Figure 2
42

Contrasting tundra responses to fertilization
Figure 3
43

Contrasting tundra responses to fertilization
Figure 4
44