bark beetle disturbance and nitrogen cycling in … · 2012-05-01 · attack (safranyik and carroll...

BARK BEETLE DISTURBANCE AND

NITROGEN CYCLING IN CONIFER FORESTS

OF GREATER YELLOWSTONE

by

Jacob M. Griffin

A dissertation submitted in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

(Zoology)

at the

UNIVERSITY OF WISCONSIN-MADISON

2011

i

TABLE OF CONTENTS

PREFACE……….………………………………………………………………………………iii

CHAPTER 1

Nitrogen cycling following mountain pine beetle disturbance in lodgepole pine

forests of Greater Yellowstone…………………………...………………………………...……1

Abstract…………………………………………………………………………………...1

Introduction………………………………………………………………………………2

Methods…………………………………………………………………………………..5

Results…………………………………………………………………………………...11

Discussion……………………………………………………………………………….16

Conclusions……………………………………………………………………………...23

Acknowledgements……………………………………………………………………..23

References……………………………………………………………………………….24

Tables……………………………………………………………………………………31

Figure legends…………………………………………………………………………..35

Figures…………………………………………………………………………………..37

CHAPTER 2

Forest type influences ecosystem response to bark beetle disturbance……………………..43

Abstract………………………………………………………………………………….43

Introduction……………………………………………………………………………..44

Methods………………………………………………………………………………….47

Results…………………………………………………………………………………...52

Discussion……………………………………………………………………………….58

ii

Conclusions.......................................................................................................................61

Acknowledgements..........................................................................................................61

References.........................................................................................................................61

Tables................................................................................................................................66

Figure legends...................................................................................................................69

Figures...............................................................................................................................71

CHAPTER 3

Does salvage logging alter ecosystem response to bark beetle disturbance

in lodgepole pine forests?............................................................................................................76

Abstract.............................................................................................................................76

Introduction......................................................................................................................77

Methods.............................................................................................................................80

Results...............................................................................................................................84

Discussion.........................................................................................................................88

Conclusions.......................................................................................................................92

Acknowledgements..........................................................................................................93

References.........................................................................................................................93

Tables................................................................................................................................99

Figure legends.................................................................................................................100

Figures.............................................................................................................................102

CONCLUSIONS........................................................................................................................110

iii

PREFACE

Disturbance has long been considered an important driver of ecosystem dynamics

(Bormann and Likens 1979, Pickett and White 1985, Chapin et al. 2002), influencing a wide

range of key ecosystem functions such as carbon storage (Li et al. 2003, Kashian et al. 2006),

nutrient retention (Pardo et al. 1995), and biodiversity (Connell 1978). The influence of

disturbance on ecosystem function may be manifested through several key mechanisms,

including the redistribution of biomass, nutrients, and energy among ecosystem pools, the

alteration of spatial pattern (Hadley 1994, Turner et al. 1999, Turner et al. 2004), and changes in

species composition (Chapin et al. 2002, Ellison et al. 2005). Effects of disturbance on

ecosystem function have been quantified for a wide range of spatial scales from small gaps

(Scharenbroch and Bockheim 2008) to whole continents (e.g.Potter et al. 2005), and often

exhibit threshold spatial extents or severities where qualitative shifts in ecosystem response

occur (Parsons et al. 1994). Temporal scale is also a key consideration, as many disturbance

types have either long-lived or delayed effects on ecosystem function (Foster et al. 1998,

Compton and Boone 2000, DeLuca and Aplet 2008). Disturbance interactions, often crossing

multiple scales, will be key to predicting future ecosystem consequences and have been

demonstrated in several systems (Paine et al. 1998, Throop and Lerdau 2004, Allen 2007, Raffa

et al. 2008).

Insect activity and insect-induced tree mortality can be significant components of the

disturbance regime in many temperate forests with substantial consequences for elemental

nutrient cycling. Mechanisms of insect disturbance involve both direct influence of herbivory

and insect-induced mortality on biomass stocks, growth rates, and ecosystem fluxes such as

litterfall and decomposition, as well as indirect effects through altered abiotic conditions or shifts

iv

in plant community composition (Mattson and Addy 1975, Schowalter 1985, Schowalter et al.

1986, Haack and Byler 1993, Hunter 2001). The great majority of studies regarding insect

disturbance and nutrient cycling have focused on folivorous insects as the study system (e.g.

Lovett and Ruesink 1995, Eshleman et al. 1998, Kosola et al. 2001, Christenson et al. 2002,

Townsend et al. 2004). Significantly less attention has been given to potential nutrient cycling

changes following outbreaks of non-folivorous insects, with notable exceptions (Huber et al.

2004, Huber 2005, Morehouse et al. 2008). By targeting specific tissues of host plants, various

insect guilds can have widely differing effects on host mortality and the transfer of materials

among ecosystem pools (Feller 2002, Lovett et al. 2006)

Even less studied is the role of forest type in determining ecosystem response to

outbreaks of the same insect guild. This knowledge gap is particularly important when

considering foundation species, whose characteristics largely determine ecosystem function

(Ellison et al. 2005) and whose dominance in forests often increases susceptibility to insect

attack (Safranyik and Carroll 2006). Variability in the baseline biogeochemistry of pure stands of

different tree species (Prescott et al. 1992, Thomas and Prescott 2000, Giardina et al. 2001,

Lovett et al. 2004) may lead to unique ecosystem responses following the same disturbance type.

Furthermore, additional compound disturbances following insect attack such as fire or salvage

logging may lead to qualitatively different effects on nutrient cycling compared to insects alone

(Paine et al. 1998, Peters et al. 2004). This is particularly important if insect outbreak affects the

probability or severity of other subsequent disturbances (Lindenmayer and Noss 2006, Page and

Jenkins 2007, Simard et al. 2011). Disturbance types vary in the specific ecosystem components

that are directly affected. For example, direct effects of bark beetles are limited to canopy

vegetation, and can be contrasted with other disturbance types such as forest fire, windstorm, or

v

logging which cause immediate change in a wider range of ecosystem components including

understory vegetation and soils (Figure 1).

Throughout western North America, several species of native bark beetles of the genus

Dendroctonus (Coleoptera: Curculionidae: Scolytinae) have been in outbreak phase over the past

decade (Raffa et al. 2008). The impacts of this disturbance type on some aspects of ecosystem

function such as carbon sequestration (Kurz et al. 2008) and hydrology (Boon 2009) are

beginning to be understood, but the consequences for nitrogen (N) cycling are less well-known.

In the Greater Yellowstone Ecosystem of WY, MT, and ID, USA, current bark beetle outbreaks

have occurred in multiple conifer forest types including lodgepole pine (Pinus contorta Dougl.),

Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), Englemann spruce (Picea engelmannii

Parry), and whitebark pine (Pinus albicaulis Engelmann). Outbreaks are allowed to progress

naturally within the National Park lands of the region, however salvage logging is being applied

as a management technique on National Forest lands to recover usable forest products and reduce

a perceived risk of increased forest fire.

Bark Beetles

Salvage logging

Bark Beetles

Salvage loggingSalvage logging

Figure 1. Impacts of bark beetle outbreak and other disturbance types on forest ecosystem components. Beetles remove large portions of the canopy, but direct impacts are limited to this component. Other disturbances including salvage logging cause additional canopy disturbance, but also may have direct impacts understory vegetation and soil. Adapted from Roberts (2004).

vi

For my dissertation I will address three general questions related to the issues described

above. (1) How do forest structure, soil microclimate, and N cycling through the litter, soil,

and vegetation change during and decades after beetle outbreak?, (2) How does host tree

forest type moderate the response of these ecosystem parameters to similar bark beetle

outbreaks?, and (3) How do forest structure, microclimate, and nitrogen cycling following

post-beetle salvage logging differ from bark beetle outbreak alone? I will approach these

questions by taking advantage of concurrent outbreaks of two Dendroctonus bark beetle species

in the Greater Yellowstone Ecosystem. First, I will use a replicated chronosequence approach to

measure the effects of mountain pine beetle (Dendroctonus ponderosae Hopkins) disturbance

over a thirty year time span in lodgepole pine forests. Second, I will explore the influence of

forest type on ecosystem response to bark beetles by comparing changes in the mountain pine

beetle/lodgepole pine system to those resulting from Douglas-fir beetle (Dendroctonus

pseudotsugae Hopkins) outbreak in Douglas-fir forests over a shorter 5 year time scale. Thirdly,

I will use a paired and replicated before-after-control-impact (BACI) experimental design in

lodgepole pine forests to compare the combined effects of mountain pine beetle outbreak and

salvage logging to the effects of bark beetles alone (Figure 2).

Figure 2. Three dissertation chapters on the influence of bark beetle disturbance on N cycling in conifer forests using multiple approaches. Chapter 1 describes a single beetle/host system over decadal scales. Chapter two compares two beetle/host systems over shorter time scales. Chapter three is an experiment testing short-term effects of the compound disturbances of beetles and salvage logging.

Chapter 1

Time scale of study

Num

ber

of sy

stem

s

Years Decades

Tw

oO

ne

Chapter 2

Influence of host forest type on ecosystem response

Chapter 1

Decadal-scale impact of bark beetles on the N cycle

Chapter 3

Compound disturbances

Bark beetles + salvage logging

Chapter 1Chapter 1Chapter 1

Time scale of study

Num

ber

of sy

stem

s

Years Decades

Tw

oO

ne

Time scale of study

Num

ber

of sy

stem

s

Years Decades

Tw

oO

ne

Chapter 2

Influence of host forest type on ecosystem response

Chapter 1

Decadal-scale impact of bark beetles on the N cycle

Chapter 3

Compound disturbances

Bark beetles + salvage logging

vii

REFERENCES Allen, C. D. 2007. Interactions across spatial scales among forest dieback, fire, and erosion in northern New Mexico landscapes. Ecosystems 10:797-808. Boon, S. 2009. Snow ablation energy balance in a dead forest stand. Hydrological Processes 23:2600-2610. Bormann, F. H., and G. E. Likens. 1979. Pattern and process in a forested ecosystem: disturbance, development, and the steady state based on the Hubbard Brook ecosystem study. Springer-Verlag, New York. Chapin, F. S., P. A. Matson, and H. A. Mooney. 2002. Principles of Terrestrial Ecosystem Ecology. Springer, New York. Christenson, L. M., G. M. Lovett, M. J. Mitchell, and P. M. Groffman. 2002. The fate of nitrogen in gypsy moth frass deposited to an oak forest floor. Oecologia 131:444-452. Compton, J. E., and R. D. Boone. 2000. Long-term impacts of agriculture on soil carbon and nitrogen in New England forests. Ecology 81:2314-2330. Connell, J. H. 1978. Diversity in Tropical Rain Forests and Coral Reefs - High Diversity of Trees and Corals Is Maintained Only in a Non-Equilibrium State. Science 199:1302-1310. DeLuca, T. H., and G. H. Aplet. 2008. Charcoal and carbon storage in forest soils of the Rocky Mountain West. Frontiers in Ecology and the Environment 6:18-24. Ellison, A. M., M. S. Bank, B. D. Clinton, E. A. Colburn, K. Elliott, C. R. Ford, D. R. Foster, B. D. Kloeppel, J. D. Knoepp, G. M. Lovett, J. Mohan, D. A. Orwig, N. L. Rodenhouse, W. V. Sobczak, K. A. Stinson, J. K. Stone, C. M. Swan, J. Thompson, B. Von Holle, and J. R. Webster. 2005. Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3:479-486. Eshleman, K. N., R. P. Morgan, J. R. Webb, F. A. Deviney, and J. N. Galloway. 1998. Temporal patterns of nitrogen leakage from mid-Appalachian forested watersheds: Role of insect defoliation. Water Resources Research 34:2005-2016. Feller, I. C. 2002. The role of herbivory by wood-boring insects in mangrove ecosystems in Belize. Oikos 97:167-176. Foster, D. R., G. Motzkin, and B. Slater. 1998. Land-use history as long-term broad-scale disturbance: Regional forest dynamics in central New England. Ecosystems 1:96-119. Giardina, C. P., M. G. Ryan, R. M. Hubbard, and D. Binkley. 2001. Tree species and soil textural controls on carbon and nitrogen mineralization rates. Soil Science Society of America Journal 65:1272-1279.

viii

Haack, R. A., and J. W. Byler. 1993. Insects and Pathogens - Regulators of Forest Ecosystems. Journal of Forestry 91:32-37. Hadley, K. S. 1994. The Role of Disturbance, Topography, and Forest Structure in the Development of a Montane Forest Landscape. Bulletin of the Torrey Botanical Club 121:47-61. Hawkins, B. J., H. Boukcim, and C. Plassard. 2008. A comparison of ammonium, nitrate and proton net fluxes along seedling roots of Douglas-fir and lodgepole pine grown and measured with different inorganic nitrogen sources. Plant Cell and Env. 31:278-287. Huber, C. 2005. Long lasting nitrate leaching after bark beetle attack in the highlands of the Bavarian Forest National Park. Journal of Environmental Quality 34:1772-1779. Huber, C., M. Baumgarten, A. Gottlein, and V. Rotter. 2004. Nitrogen turnover and nitrate leaching after bark beetle attack in mountainous spruce stands of the Bavarian Forest National Park. Water Air and Soil Pollution: Focus 4:391-414. Hunter, M. D. 2001. Insect population dynamics meets ecosystem ecology: effects of herbivory on soil nutrient dynamics. Agricultural and Forest Entomology 3:77-84. Kashian, D. M., W. H. Romme, D. B. Tinker, M. G. Turner, and M. G. Ryan. 2006. Carbon storage on landscapes with stand-replacing fires. Bioscience 56:598-606. Keane, R. E. 2008. Biophysical controls on surface fuel litterfall and decomposition in the northern Rocky Mountains, USA. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 38:1431-1445. Kosola, K. R., D. I. Dickmann, E. A. Paul, and D. Parry. 2001. Repeated insect defoliation effects on growth, nitrogen acquisition, carbohydrates, and root demography of poplars. Oecologia 129:65-74. Kurz, W. A., C. C. Dymond, G. Stinson, G. J. Rampley, E. T. Neilson, A. L. Carroll, T. Ebata, and L. Safranyik. 2008. Mountain pine beetle and forest carbon feedback to climate change. Nature 452:987-990. Li, Z., M. J. Apps, W. A. Kurz, and E. Banfield. 2003. Temporal changes of forest net primary production and net ecosystem production in west central Canada associated with natural and anthropogenic disturbances. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 33:2340-2351. Lindenmayer, D. B., and R. F. Noss. 2006. Salvage logging, ecosystem processes, and biodiversity conservation. Conservation Biology 20:949-958.

ix

Lovett, G. M., C. D. Canham, M. A. Arthur, K. C. Weathers, and R. D. Fitzhugh. 2006. Forest ecosystem responses to exotic pests and pathogens in eastern North America. Bioscience 56:395-405. Lovett, G. M., and A. E. Ruesink. 1995. Carbon and Nitrogen Mineralization from Decomposing Gypsy-Moth Frass. Oecologia 104:133-138. Lovett, G. M., K. C. Weathers, M. A. Arthur, and J. C. Schultz. 2004. Nitrogen cycling in a northern hardwood forest: Do species matter? Biogeochemistry 67:289-308. Mattson, W. J., and N. D. Addy. 1975. Phytophagous Insects as Regulators of Forest Primary Production. Science 190:515-522. Morehouse, K., T. Johns, J. Kaye, and A. Kaye. 2008. Carbon and nitrogen cycling immediately following bark beetle outbreaks in southwestern ponderosa pine forests. Forest Ecology and Management 255:2698-2708. Page, W. G., and M. J. Jenkins. 2007. Mountain pine beetle-induced changes to selected lodgepole pine fuel complexes within the intermountain region. For Sci 53:507-518. Paine, R. T., M. J. Tegner, and E. A. Johnson. 1998. Compounded perturbations yield ecological surprises. Ecosystems 1:535-545. Pardo, L. H., C. T. Driscoll, and G. E. Likens. 1995. Patterns of nitrate loss from a chronosequence of clear-cut watersheds. Water Air and Soil Pollution 85:1659-1664. Parsons, W. F. J., D. H. Knight, and S. L. Miller. 1994. Root Gap Dynamics in Lodgepole Pine Forest : Nitrogen Transformations in Gaps of Different Size. Ecol App 4:354-362. Peters, D. P. C., R. A. Pielke, B. T. Bestelmeyer, C. D. Allen, S. Munson-McGee, and K. M. Havstad. 2004. Cross-scale interactions, nonlinearities, and forecasting catastrophic events. Proceedings of the National Academy of Sciences of the United States of America 101:15130-15135. Pickett, S. T. A., and P. S. White, editors. 1985. The ecology of natural disturbance and patch dynamics. Academic Press, Orlando, FL. Potter, C., P. N. Tan, V. Kumar, C. Kucharik, S. Klooster, V. Genovese, W. Cohen, and S. Healey. 2005. Recent history of large-scale ecosystem disturbances in North America derived from the AVHRR satellite record. Ecosystems 8:808-824. Prescott, C. E., J. P. Corbin, and D. Parkinson. 1992. Availability of Nitrogen and Phosphorus in the Forest Floors of Rocky-Mountain Coniferous Forests. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 22:593-600.

x

Raffa, K. F., B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H. Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic amplification: The dynamics of bark beetle eruptions. Bioscience 58:501-517. Roberts, M. R. 2004. Response of the herbaceous layer to natural disturbance in North American forests. Canadian Journal of Botany-Revue Canadienne De Botanique 82:1273-1283. Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine beetle in lodgepole pine forests. Pages 3-66 in L. Safranyik and B. Wilson, editors. The mountain pine beetle: A synthesis of biology, management, and impacts on lodgepole pine. Nat. Res. Can., Can. Forest Service, Pacific Forestry Centre, Victoria, BC. Scharenbroch, B. C., and J. G. Bockheim. 2008. The effects of gap disturbance on nitrogen cycling and retention in late-successional northern hardwood-hemlock forests. Biogeochemistry 87:231-245. Schowalter, T. D. 1985. Adaptations of Insects to Disturbance. Pages 235-252 in S. T. A. Pickett and P. S. White, editors. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando. Schowalter, T. D., W. W. Hargrove, and D. A. Crossley. 1986. Herbivory in Forested Ecosystems. Annual Review of Entomology 31:177-196. Simard, M., W. H. Romme, J. M. Griffin, and M. G. Turner. 2011. Do mountain pine beetle outbreaks change the probability of active crown fire in lodgepole pine forests? Ecological Monographs 81:3-24. Thomas, K. D., and C. E. Prescott. 2000. Nitrogen availability in forest floors of three tree species on the same site: the role of litter quality. Canadian Journal of Forest Research- Revue Canadienne De Recherche Forestiere 30:1698-1706. Throop, H. L., and M. T. Lerdau. 2004. Effects of nitrogen deposition on insect herbivory: Implications for community and ecosystem processes. Ecosystems 7:109-133. Townsend, P. A., K. N. Eshleman, and C. Welcker. 2004. Remote sensing of gypsy moth defoliation to assess variations in stream nitrogen concentrations. Ecological Applications 14:504-516. Turner, M. G., W. H. Romme, and R. H. Gardner. 1999. Prefire heterogeneity, fire severity, and early postfire plant reestablishment in subalpine forests of Yellowstone National Park, Wyoming. International Journal of Wildland Fire 9:21-36. Turner, M. G., D. B. Tinker, W. H. Romme, D. M. Kashian, and C. M. Litton. 2004. Landscape patterns of sapling density, leaf area, and aboveground net primary production in postfire lodgepole pine forests, Yellowstone National Park (USA). Ecosystems 7:751-775.

1

CHAPTER 1

Nitrogen cycling following mountain pine beetle disturbance in

lodgepole pine forests of Greater Yellowstone

ABSTRACT

Widespread bark beetle outbreaks are currently affecting multiple conifer forest types

throughout western North America, yet many ecosystem-level consequences of this disturbance

are poorly understood. We quantified the effect of mountain pine beetle (Dendroctonus

ponderosae) outbreak on nitrogen (N) cycling through litter, soil, and vegetation in lodgepole

pine (Pinus contorta var. latifolia) forests of the Greater Yellowstone Ecosystem (Wyoming,

USA) across a 0-30 year chronosequence of time-since-beetle disturbance. Recent (1-4 years)

bark beetle disturbance increased total litter depth and N concentration in needle litter relative to

undisturbed stands, and soils in recently disturbed stands were cooler with greater rates of net N

mineralization and nitrification than undisturbed sites. Thirty years after beetle outbreak, needle

litter N concentration remained elevated; however total litter N concentration, total litter mass,

and soil N pools and fluxes were not different from undisturbed stands. Canopy N pool size

declined 58% in recent outbreaks, and remained 48% lower than undisturbed in 30-year old

outbreaks. Foliar N concentrations in unattacked lodgepole pine trees and an understory sedge

were positively correlated with net N mineralization in soils across the chronosequence. Bark

beetle disturbance altered N cycling through the litter, soil, and vegetation of lodgepole pine

forests, but changes in soil N cycling were less severe than those observed following stand

replacing fire. Several lines of evidence suggest the potential for N leaching is low following

bark beetle disturbance in lodgepole pine.

2

INTRODUCTION

Bark beetle disturbance and nitrogen cycling

Disturbance regulates a wide range of ecosystem properties (Bormann and Likens 1979,

Pickett and White 1985, Chapin et al. 2002), and in forests, disturbance-induced tree mortality

can change stand structure, productivity, and nutrient cycling (Lodge et al. 1994, Ostertag et al.

2003). Because canopy foliage stores nutrients, determines litter quantity and quality, and

moderates litter-soil microclimate (Prescott 2002), canopy disturbance may influence soil

nutrient availability (Chapin et al. 2002). Previous studies of disturbance and nutrient cycling in

forest ecosystems have provided insight into the magnitude, timing and mechanisms of response

(e.g.Vitousek et al. 1979). However, disturbances vary spatially and temporally in both pattern

and severity, and ecosystem response varies among disturbance types and forest communities.

Insect outbreaks are common disturbances in forest ecosystems, though most studies of

insect herbivores and nutrient cycling have addressed Lepidopteran (e.g. Lovett and Ruesink

1995, Kosola et al. 2001, Christenson et al. 2002, Houle et al. 2009) and Homopteran (e.g.

Kizlinski et al. 2002, Stadler et al. 2006) insects. Few studies have focused on stem phloem-

feeding insects, such as Dendroctonus and Ips bark beetle species (Coleoptera: Curculionidae-

Scolytinae). In contrast to folivores, bark beetles kill trees by consuming cambium and

disrupting phloem flow, thereby acting as agents of selective mortality within a forest. Because

bark beetle outbreaks have reached unprecedented levels throughout western North America

(Raffa et al. 2008), understanding the ecological effects of widespread beetle-induced tree

mortality has become increasingly important.

Bark beetles are native to temperate and boreal coniferous forests and intermittent

outbreaks typically recur at decadal-scale intervals (Raffa et al. 2008). Successful beetle attack in

3

summer causes tree death by the following spring, when needles turn red and begin to fall; the

stand progresses from this “red stage” to the “gray stage” as all needles are shed (Safranyik and

Carroll 2006). Large trees are most susceptible to bark beetle attack, and mortality of large trees

can reach 100% (Safranyik and Carroll 2006). When the outbreak subsides, release of subcanopy

surviving trees is often the major mechanism of regeneration (Romme et al. 1986, Nigh et al.

2008, Vyse et al. 2009). Herbaceous species and shrubs also increase presumably in response to

light and water availability (McCambridge et al. 1982, Stone and Wolfe 1996), but increased

nitrogen availability may play a role. The effects of bark beetles on stand structure (tree density,

basal area, and species composition) have been well documented (Shore et al. 2006, Dordel et al.

2008, Klutsch et al. 2009). However, the consequences of bark beetle outbreaks for nitrogen (N)

cycling have not been widely studied.

The effect of bark beetle outbreak on N pools and fluxes depends on the balance between

factors that enhance or limit nutrient supply. Stand-level transpiration declines as trees within the

stand begin to die, and soil moisture increases. Nutrient uptake by trees also declines, but soil

nutrient pools will increase only if the rate of nutrient supply via mineralization exceeds the rate

of nutrient removal via uptake and leaching. A post-outbreak pulse of litter could increase soil

inorganic N pools (Cullings et al. 2003), but conifer litter can also immobilize soil N (Fahey et

al. 1985). Canopy and litter changes also affect incident radiation (Hais and Kucera 2008), air

flow (Boon 2009) and soil insulation (Byers 1984), and experimental additions of lodgepole pine

litter have been shown to modify litter-soil microclimate and soil N availability (Cullings et al

2003). In an N-saturated Norway spruce (Picea abies (L.) Karst.) forest in Bavaria, Germany,

soil nitrate concentrations were elevated for 7 years after an outbreak of Ips typographus, but

returned to pre-outbreak levels after 17 years (Huber 2005). In soils of southwestern US Pinus

4

ponderosa forests, disturbance by Ips and Dendroctonus bark beetle species increased soil

ammonium and laboratory net nitrification rates, but did not affect soil nitrate or laboratory net

mineralization rates (Morehouse et al. 2008). Additional studies that compare a broader range of

N pools and fluxes, both during and after an outbreak, with unaffected stands are needed to

understand how bark beetle disturbance alters N flow through the litter, soil, and vegetation over

time.

Approach and hypotheses

We studied a 0-30 year chronosequence of mountain pine beetle (MPB; Dendroctonus

ponderosae Hopkins) disturbance in lodgepole pine (Pinus contorta var. latifolia Dougl.) forests

of the Greater Yellowstone Ecosystem. Specifically, we asked how litter quantity and quality,

soil nitrogen pool and fluxes, and foliar nitrogen varied with time-since-beetle outbreak and

evaluated several hypotheses. We expected litter depth, litter mass, and total litter N

concentration to increase during an outbreak and decrease to below undisturbed levels 30 years

after an outbreak. As observed by Morehouse et al. (2008) for Pinus ponderosa, we expected N

concentrations in fresh needle litter to be elevated during the outbreak due to lack of N resorption

prior to litterfall. For soil inorganic N, we expected nitrate and ammonium pools to increase

during an outbreak but not differ from undisturbed stands by 30 years after an outbreak..

Expectations for net N mineralization and net nitrification during the outbreak were less clear;

abiotic changes may stimulate mineralization rates, but microbial populations, surviving

vegetation and a pulse of needle litter could also act to immobilize soil N. By 30 years after an

outbreak, however, we expected N mineralization to be lower than undisturbed stands because

soils would likely be warmer and drier under an open canopy, and litter inputs would be

diminished relative to undisturbed stands since canopy biomass is not yet recovered. Falling

5

beetle-killed snags may also reduce soil N availability and turnover during this period because

coarse wood may be a net N sink (Fahey et al. 1985, Busse 1994, Laiho and Prescott 1999) and

can reduce decomposition and mineralization by altering microsite conditions (Remsburg and

Turner 2006, Metzger et al. 2008). For foliar nitrogen in lodgepole pine, we expected the foliar N

concentration of surviving trees to increase during an outbreak and be positively correlated with

net N mineralization, but total foliar N pool to decline with tree mortality. We further expected

that these differences would persist 30 years after an outbreak because canopy biomass would

still be greatly reduced relative to undisturbed stands, increasing resource availability per tree

even if soil N pools and fluxes return to or fall below undisturbed levels. In the understory sedge

Carex geyerii, we similarly predicted foliar N concentration to increase during an outbreak and

be positively correlated with N mineralization.

METHODS

Study area

This study was conducted in subalpine forests of the Greater Yellowstone Ecosystem

within Yellowstone National Park and Bridger-Teton National Forest in northwestern Wyoming,

USA. Lodgepole pine is a common forest type of the region, and most soils are nutrient-poor and

derived from volcanic rhyolitic and andesitic deposits. Climate is characterized by cool winters

and dry summers, with mean July temperatures of 12.8 ºC and mean January temperatures of -

11.7 ºC at Yellowstone Lake (WRCC 2010b). Mean annual precipitation is 563 mm, mostly

occurring as snow, and increases with elevation (Dirks and Martner 1982). Extensive outbreaks

of the MPB occurred in southern and western Yellowstone National Park in the 1970s and 1980s

(Lynch et al. 2006). Since 2003, large areas of forest in the eastern and southern portions of

6

Greater Yellowstone have been affected by outbreaks of MPB in lodgepole pine and whitebark

pine (Pinus albicaulis Engelmann), Douglas-fir beetle (Dendroctonus pseudotsugae Hopkins) in

Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), and spruce beetle (Dendroctonus rufipennis

Kirby) in Engelmann spruce (Picea engelmannii Parry). This study focused on lodgepole pine

forests affected by MPB in both the current outbreak and previous outbreaks of the 1970’s and

1980’s.

Sampling design

We sampled a time-since-beetle (TSB) chronosequence including four classes with five

replicates each (N = 20 plots). TSB classes included undisturbed stands; stands within the current

outbreak, both red stage (2 years post-outbreak) and gray stage (4 years post-outbreak); and

stands attacked by bark beetles approximately 30 years ago. Potential sites were identified using

maps of current (USFS 2006) and historic (Lynch et al. 2006) mountain pine beetle outbreaks,

soil type, and forest age. Field inspection assured selection of stands with comparable species

composition, basal area, soils, and bark beetle mortality. Undisturbed and 30-year TSB stands

were located in Yellowstone National Park, while red and gray stands were located within the

current outbreak on the Bridger Teton National Forest (Figure 1). Disturbed plots were located

within stands (0.25 ha) of homogenous structure and beetle disturbance intensity; undisturbed

sites were located in comparable beetle-susceptible stands. In lodgepole pine, both beetle

susceptibility (Safranyik and Carroll 2006) and ecosystem N status (Smithwick et al. 2009) are

largely determined by stand structure and age. Thus, site selection based upon stand structure,

age, and substrate largely controlled for pre-disturbance differences in these factors among

classes. To validate the chronosequence, dendrochronological analyses were used to determine

post-fire stand age for all plots and to reconstruct pre-outbreak stand structure and MPB outbreak

7

severity for the 30 year TSB plots (Simard et al. 2011). An 8-m radius plot (201 m2) was

established at each site in summer 2007, and slope and aspect were recorded at the plot center.

Microclimate

To evaluate variation in growing-season microclimate, soil and air temperature were

measured hourly in three of the five plots within each TSB class (N = 12 instrumented plots)

from June 18 through August 7, 2008, using three pairs of iButton datalogger probes (Maxim

Integrated Products Inc., Dallas Semiconductor, Sunnyvale, CA) per plot. One iButton of each

pair was installed at the litter-organic soil interface, and the second was installed 10 cm below

this interface. Air temperature was recorded using another temperature probe hung inside a

covered but well-ventilated and open-bottomed white PVC housing attached to a tree at breast

height near the plot center.

For each soil depth, temperature probe data were summarized as follows. First, plot-

level hourly temperatures were determined by averaging data from the three probes per plot. To

account for geographic variation and normalize comparisons among plots, hourly differences

between air and soil temperature were calculated (difference = soil – air). Plot-level daily mean

temperatures, temperature ranges, and soil-air temperature differences were then calculated from

the hourly data, and averaged by class to determine class-level daily means. Plot-level growing

season mean temperatures, ranges, and soil-air temperature differences were calculated by

averaging daily means across the sampling period, and class-level growing season means were

then calculated from these plot-level data.

Vegetation

All live and dead standing trees > 1.4 m tall were identified to species and measured for

diameter at breast height (DBH). MPB-killed trees were identified by the presence of pitch tubes

8

(tree resin accumulation at boring hole entrances), boring dust, and J-shaped galleries under the

bark (diagnostic of Dendroctonus (Safranyik and Carroll 2006)). Downed logs rooted in the plot

were also measured for DBH, identified to species, and scored as MPB-killed if J-shaped

galleries were visible. Trees < 1.4 m tall were measured by 10 cm height classes in the northeast

quadrant of the plot (50m2). Pre-outbreak basal area for red and gray stands was calculated by

summing the basal area of live and beetle-killed trees. In the 30-year TSB stands post-outbreak

growth of survivors was substantial, so pre-outbreak basal area was determined from

dendrochronology using stand reconstruction techniques (Simard et al 2011). Beetle-killed basal

area in the 30-year TSB stands was determined by summing the basal area of downed logs with

evidence of MPB galleries. Lodgepole pine canopy biomass was calculated for each plot using

allometrics developed for this region (Brown 1978). Ground cover was visually estimated to the

nearest 10% by plant functional groups (forbs, sedges, grasses, and seedlings) in ten 0.25-m2

circular microplots. The microplots were located within the inner 5-m plot radius using a

stratified random design of fixed distances (one at 0.5-m; two at 1.5-m, 2.5-m and 3.5-m; and

three at 4.5-m) and random bearings (in 10º increments) from the plot center, and averaged by

plot. Biomass of the sedge Carex geyerii was calculated from allometrics previously reported for

Yellowstone National Park (Turner et al. 2004).

Litter quantity and quality

In each 0.25-m2 microplot, litter depth was recorded at three locations and a 400-cm2

sample of the litter layer was collected and oven-dried at 60 ºC. Plot-level litter depth and mass

were obtained by averaging values from the 10 microplots. Litter from each microplot was sorted

into three categories: fresh current-year needle litter, identified by bright red color and lack of

mottling on surface (Morehouse et al. 2008); all-needle litter (all ages of needles combined); and

9

total litter (all foliar litter components and woody litter components < 1.0 cm wide). Sorted litter

was composited by plot for each litter category, and ground to powder for C:N analysis on a

Leco CNS-2000 at the University of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL

2010)

Soil chemistry, N pools, and N fluxes

One soil core was collected from each 0.25-m2 microplot using a 5-cm diameter x 15-cm

long PVC corer. Soils were sieved (2 mm), weighed, and divided into three subsamples: 30 g

oven-dried at 60 ºC for gravimetric percent moisture; 20 g extracted in 75 ml of 2M KCl for 2

hours, with the extract then filtered and frozen for later analysis of NH4+ and NO3

- pools; and 20

g air-dried and bulked by plot for soil texture and chemical analyses. Air-dried soil was analyzed

for pH, total N, exchangeable Ca, Mg, and K, available P (Bray P1 extract), and organic matter

content at the University of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL 2010).

Soil organic N was determined by difference using total N and inorganic N values, and soil

texture was determined using the Bouyoucos hydrometer technique (Bouyoucos 1962).

Net N mineralization and net nitrification were measured using ion-exchange resin cores

(Binkley et al. 1992), with one core incubated in situ for approximately one year (July 2007 -

June 2008) in each 0.25-m2 microplot. Incubated cores consisted of a 5-cm diameter x 15-cm

long PVC tube of soil with an ion-exchange resin bag placed at the bottom. Resin bags were

constructed using 20 g of mixed bed ion exchange resin (J.T Baker #JT4631-1) tied inside a

piece of un-dyed nylon stocking material. Upon retrieval in summer 2008, core soils were sieved

(2 mm) and core soils and resin bags were extracted separately in 2M KCl in the same manner as

the 2007 initial soil samples described above. All KCl extractions were analyzed for [NH4+] and

10

[NO3-] using colorimetric methods on an Astoria Pacific II continuous flow autoanalyzer. For

each microplot, mineralization and nitrification rates were calculated as:

rate = [(final soil N + resin bag N) – initial soil N ] / incubation time

and expressed in units of µg N g soil-1 year-1. Values were then averaged by plot. Atmospheric N

inputs in the region are low (~ 1 kg N ha-1 year-1; and were not factored into the soil N

calculations. (www.epa.gov/castnet/charts/YEL408_totn.png)

Foliar N

In each plot, pole pruners were used to collect canopy foliage from three live mature

lodgepole pines unattacked by MPB. Two fully sunlit branches (0.5-m long) were clipped from

each tree and separated into two subsamples: current-year foliage (from one branch per tree), and

all-years foliage (from the second branch per tree). Understory sedge (Carex geyerii Boott)

foliage was collected by clipping 15-20 pieces from each quadrant of the plot. For both

lodgepole and sedge foliar N, samples from two of the five undisturbed sites were excluded from

the analyses reported here because they were sampled much earlier in the growing-season than

the remaining eighteen sites. All foliar samples were oven-dried at 60ºC and ground to powder

for C and N analysis on a Leco CNS-2000 analyzer at the University of Wisconsin Soil and Plant

Analysis Lab. Tree-level canopy data (N=3 per sample type) and quadrant-level sedge data (N

=4) were averaged by plot. To compute canopy and Carex geyerii N pool sizes, the mean foliar

N concentration for each plot was multiplied by biomass values derived from the allometrics

cited above.

Statistical analyses

All statistical analyses were performed in SAS (SAS Institute Inc. 2003), and unless

otherwise noted all reported variance values are two standard errors. ANOVA was used to test

11

for differences among TSB classes in stand basal area, site characteristics (slope, elevation,

aspect, and soil chemistry), understory cover, soil temperatures, litter depth, litter quantity and

quality, soil N pools, foliar N concentrations, and foliar N pools. When ANOVAs were

significant, Tukey’s HSD test (α = 0.05) was used to identify differences among means.

Because variability in soil characteristics of P. contorta forests in this region is known to

influence nitrogen transformations irrespective of bark beetle outbreaks (Smithwick et al. 2005),

we used general linear models to account for potential covariate effects (soil cations, C, OM, pH,

bulk density, and C:N ratio) when testing for differences in net mineralization and net

nitrification rates among TSB classes. To explore which beetle-induced changes were related to

differences in soil N cycling rates among TSB classes, we used linear regression to identify

relationships among litter, soil temperature, and N mineralization variables. Because only three

out of five plots per TSB class were instrumented with soil temperature probes, regressions

including soil temperature variables have N = 12 rather than N = 20. We also used linear

regression to test whether foliar N concentrations in unattacked P. contorta and C. geyerii were

positively related to N net mineralization rates.

RESULTS

Stand structure and site characteristics

Pre-disturbance stand structure and disturbance severity were similar across the

chronosequence. There were no significant differences among TSB classes in pre-outbreak post-

fire stand age (168 ± 18 years), pre-outbreak P. contorta basal area (43.4 ± 2.4 m2 ha-1 ),

background P. contorta mortality (7.2 ± 1.1 m2 ha-1), or beetle-killed P. contorta basal area

12

(disturbed classes only; 33.7 ± 3.2 m2 ha-1) (Simard et al. 2011). In the 30-year TSB sites, peak

mortality occurred 29 ± 2 years prior to sampling (Simard et al. 2011).

Topographic position (elevation, aspect, and slope) did not vary among undisturbed, red,

and gray stands (Table 1). However, the 30-year TSB sites were on average 260 m lower in

elevation than other classes and had shallower slopes than gray sites (Table 1). Soil texture was

similar among TSB classes, with only percent sand and bulk density being slightly lower in the

30-year TSB compared to gray stands (Table 1). Soil K (average 160 ± 9 ppm), P (average 18 ±

2 ppm), and percent organic matter (average 4.5 ± 0.5) did not differ among TSB classes (Table

1). Soil pH, calcium and magnesium differed with the regional distribution of sites, with slightly

higher pH (0.5 pH units) and 2-3 fold higher [Ca] and [Mg] in the Bridger Teton National Forest

sites (red and gray) than in the Yellowstone National Park sites (undisturbed and 30-year TSB)

(Table 1).

Total percent cover of understory vegetation (F = 4.49, R2 = 0.46, P = 0.02) and percent

cover of forbs (F = 7.06, R2 = 0.57, P = 0.003) varied with TSB class (Figure 2). Total

understory cover in the red class averaged 60%, approximately double that of undisturbed forest;

understory cover in the gray and 30-year TSB classes was intermediate (Figure 2). Forb cover

averaged 25% in the red class, approximately 2.5x greater than in undisturbed forest (Figure 2).

Grass, sedge and shrub cover did not vary with TSB class (P > 0.05).

Microclimate

Air and soil temperatures during the 2008 growing season differed among TSB classes,

and the 30-year TSB class was both the warmest and the most variable (Table 2). Mean daily air

temperature in the 30-year TSB class was ~1 ˚C warmer than undisturbed, red, and gray classes,

and the mean daily range of temperature extremes was greater (Table 2). Temperatures at the

13

litter-soil interface varied more among classes than did air temperatures, and were 4-6 ˚C warmer

in the 30-year TSB class compared to undisturbed, red, and gray. Temperatures at the litter-soil

interface did not differ from air temperature in the undisturbed class, but were ~2˚C cooler than

air temperature in the red and gray classes, and 3˚C warmer than air temperature in the 30-year

TSB class (Table 2). At 10 cm soil depth, temperatures were 3-4˚C warmer in the 30-year TSB

class compared to the undisturbed, red, and gray. Soil temperatures at 10 cm depth were always

cooler than air temperatures but were 5.3 ± 0.2 ˚C less than air in undisturbed, red, and gray

sites, and only 2.5 ± 0.9˚C less than air in the 30-year TSB class (Table 2). There were no

differences among classes in the slope or aspect of instrumented sites (Table 2). Mean elevation

of the instrumented undisturbed sites was not significantly different from the instrumented sites

in any other class (Table 2). Thus, differences in soil temperature between undisturbed and any

disturbance class suggest an effect of TSB rather than topographic differences among classes.


Total litter depth was approximately 3 cm in red and gray stands, about twice that of

undisturbed and 30-year TSB stands (Figure 3a). However, total litter mass averaged 1577 ± 149

g m-2 and did not vary with TSB class (Figure 3b). The N concentration in fresh current- year-

needle litter (Figure 3c) and all-needle litter (Figure 3d) varied with TSB class, with the highest

values (~0.75%) observed in the red and gray classes and lower concentrations (~0.4-0.5%) in

the undisturbed and 30-year TSB classes. However, N concentration of total litter averaged 0.78

± 0.02% and did not vary with TSB class (Figure 3e). Litter N pool size averaged 13.8 ± 1.2 g N

m-2 in undisturbed, red and gray stands and was significantly lower (7.4 ± 2.5 g N m-2) in the 30-

year TSB class compared to the gray (Figure 3f).

14

Soil N pools and fluxes

Among soil N pools, only soil ammonium pool size varied with TSB class (Figure 4a).

Extractable NH4+ was six times greater in the red stands (4.75 µg N g soil-1) compared to the

undisturbed stands (0.74 µg N g soil-1) and intermediate in the gray and 30-year TSB stands.

Extractable NO3- was low (averaging 0.63 ± 0.20 µg N g soil-1) and did not vary with TSB class

(Figure 4b). Total extractable inorganic N averaged 3.11 ± 0.63 µg N g soil-1 (data not shown)

and also did not vary with TSB class, nor did soil C:N ratio, total percent soil N, or soil organic

N pool (Table 1).

Annual net N mineralization rate was approximately double (21 ± 7 µg N g soil-1 year-1)

in the red and gray stages relative to the undisturbed and 30-year TSB classes (Figure 4c). After

accounting for variation due to soil bulk density, pH and C:N ratio, the effect of TSB class on net

N mineralization was significant; collectively, these variables explained 70% of the variation in

net N mineralization (Table 3). Similar to annual net N mineralization, annual net nitrification

rate was highest in the red stage (8.2 ± 4.2 µg N g soil-1 year-1) and about three times greater than

in the undisturbed and 30-year TSB stages (Figure 4d). After accounting for soil bulk density,

there was a significant effect of TSB class on annual net nitrification (Table 3). Net N

mineralization and net nitrification were positively correlated (R2 = 0.66, P < 0.0001), and

nitrification fraction (ratio of nitrification to mineralization) averaged 0.32 ± 0.05 and did not

vary among TSB classes (data not shown). In post-incubation soils, neither nitrate concentration

(average 2.23 ± 0.67 µg N g soil-1) nor the ratio of nitrate to total inorganic N (average 0.29 ±

0.07) differed among classes (P = 0.1314 and P = 0.1710, respectively; data not shown).

Net nitrogen mineralization rates were more strongly correlated with soil temperature

(Figure 5b; R2=0.39) than with litter quality variables (Figure 5c-f; R2 <0.19). Increased litter

15

depth in the red and gray classes corresponded to decreased soil temperatures, and shallow litter

layers in the 30-year TSB class were associated higher soil temperatures (Figure 5a). Net N

mineralization rates declined with increasing soil temperature across the TSB chronosequence

(Figure 5b). Net N mineralization was weakly related to the nitrogen concentration in fresh

needlefall (Figure 5c), but showed no relationship to the nitrogen concentration in all-needle

litter or total litter, or to the total litter N pool size (Figures 5d-f).

Foliar biomass and N

Foliar biomass of live lodgepole pines was 69% lower in the current outbreak (red and

gray sites) relative to undisturbed sites, and 48% lower in the 30-year TSB class (Table 4). Pre-

outbreak foliar biomass of beetle-killed trees in the red and gray classes averaged 687 ± 49 g m-2

(data not shown), which also represents the total mass of needle litterfall induced by the

outbreak. The N concentration in current year foliage of unattacked P. contorta averaged 0.99 ±

0.02% in the red, gray and 30-year TSB classes compared to 0.82% in the undisturbed, and foliar

C:N was concomitantly lower in the disturbed classes compared to undisturbed forest (Table 4).

The N concentration in all-age foliage of unattacked P. contorta was highest in the red and gray

classes and lower in the undisturbed and 30-year TSB classes (Table 4). P. contorta foliar N pool

averaged 8.3 g N m-2 in undisturbed forest, was about 64% lower in the red and gray classes, and

48% lower in the 30-year TSB class (Table 4). Across all TSB classes, the N concentration of

current year P. contorta foliage was positively related to soil N mineralization (Figure 6a).

Biomass of the understory sedge Carex geyerii did not differ across the chronosequence (Table

4), but foliar N concentration of C. geyerii was 50% greater (1.5-1.6% N) in red and gray stands

compared to undisturbed and 30-year TSB stands. Foliar N concentration of Carex geyerii was

also positively related to annual net N mineralization across all TSB classes (Figure 6b).

16

DISCUSSION

This study demonstrates the substantial effects of MPB outbreak on N cycling through

the litter, soils, and vegetation of lodgepole pine forests across a replicated chronosequence of

time-since-beetle disturbance. Beetle-induced tree mortality triggered a cascade of effects

through increased needlefall and reduced N uptake, which in turn altered soil microclimate,

increased soil N availability, and increased N concentration of surviving vegetation in recent

outbreaks. Thirty years after the disturbance soil N parameters returned to undisturbed levels,

though changes in soil temperature, %N of new foliage, canopy biomass, canopy N pool size;

and the % N of needle litter were still evident. Our results also suggest several mechanisms that

contribute to N retention in beetle-affected lodgepole pine forests and make N loss from the

system unlikely.

Microclimate

A strong and persistent effect of bark beetle outbreak was observed on soil microclimate

during the growing season. Within the current outbreak, soils were notably cooler than in

undisturbed sites, most likely in response to increased litter depth. Though soil moisture was not

measured in this study, soil moisture probably increased due to deeper litter and reduced

transpiration caused by tree mortality (Stottlemyer and Troendle 1999, Tan et al. 2008, Griffiths

et al. 2010). Experimental additions of needle litter have been shown to increase soil moisture in

lodgepole pine forests even in the absence of tree mortality (Cullings et al. 2003). In the 30-year

TSB sites soils were warmer and likely drier than in undisturbed forest, which can decrease

decomposition (Remsburg and Turner 2006) and net N mineralization (Metzger et al. 2008) rates

in post-fire lodgepole pine forests. Though we did not measure soil moisture or decomposition

17

directly, our results suggest similar mechanisms may be important in beetle-disturbed lodgepole

pine forests as well.


The elevated N concentration of fresh needle litter in the current outbreak was consistent

with our hypothesis and prior studies, and is likely due to a lack of N resorption prior to needle

drop (Morehouse et al. 2008). However, we were surprised that the mass, N concentration, and

N pool size of the total litter layer did not differ as expected. Gradual inputs of litterfall over

several years combined with ongoing litter decomposition may explain why litter mass did not

increase significantly. Foliar mass of beetle-killed trees in red and gray stands was estimated to

be 687 ± 49 g m-2 (data not shown), which is approximately half the mass of litter observed in

undisturbed stands. Thus, if all beetle-killed foliage fell at once one may expect a 50% increase

in the litter mass of attacked stands. However, beetle-induced litterfall extends over 3-4 years,

and P. contorta litter can lose 30% of its mass after two years (Remsburg and Turner 2006).

These processes may explain why the mass gained in the litter layer is smaller than the mass lost

from the canopy by ~200 g m-2, and why increases in litter mass following beetle outbreak (498

g m-2) are small relative to the amount of litter already present on the forest floor prior to

outbreak (1512 ± 457 g m-2).

Thirty years following outbreak, we also did not see the hypothesized declines in total

litter biomass, N concentration or N pool size relative to undisturbed forest, even though stand

basal area and therefore fresh litter inputs had not yet recovered. We suggest that persistent

warming and drying of the forest floor and drier soil conditions under elevated coarse wood

(Remsburg and Turner 2006) may have reduced decomposition rates during the post-outbreak

period enabling more litter accumulation despite reduced input rates. The elevated N

18

concentration in all-aged needle litter also may represent a legacy effect of outbreak-induced

litterfall undergoing an extended period of decomposition and N immobilization.

Soil N pools and fluxes

Although we expected both soil NH4+ and NO3

- to increase during the outbreak, only soil

NH4+ was elevated despite observed increases both net N mineralization and net nitrification.

Increased soil NO3- has been observed following bark beetle outbreaks in other forest types

(Huber 2005), however several studies of lodgepole pine have shown that mortality must be

extensive before soil NO3- increases. Parsons et al. (1994) found elevated NO3

- only in large

experimental root gaps in which 30 trees were killed, and Knight et al. (1991) found no elevated

NO3- in soil solution following a 60% thinning but a 10-40 fold increase following clearcut.

Plot-level tree mortality in this study averaged 76%, but surviving trees were intermixed with

beetle-killed trees. Thus, undisturbed vegetation and microbial biomass were likely able to

assimilate the products of increased nitrification yielding no change in NO3- pool size. In contrast

to many N-saturated forests of Europe and Eastern North America where atmospheric N

deposition is high and nitrate levels can rise dramatically following disturbance (Huber 2005;

Bormann and Likens 1979), conifer forests of western North America receive little N deposition

and often show little response of nitrate following disturbance (Prescott et al. 2003; Titus et al.

2006).

Observed rates of net N mineralization and net nitrification in the undisturbed stands are

within ranges reported by other studies from Greater Yellowstone for both mature (Turner et al.

2007) and 15 year-old post-fire (Metzger et al. 2008) lodgepole pine forests. Nitrification is

typically low (<2 µg N g soil-1 year-1) in undisturbed lodgepole pine (Turner et al. 2007), and

despite a significant increase in the red stage, net nitrification was never above 8.2 µg N g soil-1

19

year-1 in any class. The magnitude of increased mineralization in the current outbreak is

comparable to that reported by Parsons et al. (1994) for artificial root gaps of 15 lodgepole pine

trees two years after disturbance, but less than increases reported for silvicultural gaps in

Douglas-fir forests 6-8 years after disturbance (Thiel and Perakis 2009). Though we did not see

any change in net N mineralization in the 30-year TSB stands relative to the undisturbed stands,

decadal-scale effects of lodgepole pine mortality on net N mineralization have been documented

in other studies. Twenty-four years following treatment, Tan et al. (2008) observed twice the

rate of N mineralization in lodgepole stands thinned to 1600 stems ha-1, compared to unthinned

stands. However, stands in that study were much younger (22 years) at the time of disturbance

than ours (average 168 years), soils were generally more fertile (e.g., soil C:N ratios in control

and thinned plots ranged from 20-33), and thinning residues were removed.

Foliar N and biomass

Nitrogen concentration of new lodgepole pine foliage demonstrated a rapid response of

unattacked trees to increased nutrient availability. The foliar N concentration of current-year

needles was positively correlated with net N mineralization, and lodgepole pine is known to take

up both NH4+ and NO3

- (Hawkins et al. 2008). Fertilization studies in lodgepole pine have also

demonstrated rapid response of foliar N to increased soil N availability, though this may

attenuate after several years as foliage biomass increases (Blevins et al. 2005). Increased foliar N

concentration in unattacked trees provided a mechanism for N retention in the canopy following

bark beetle disturbance. Between the undisturbed and gray classes, beetle outbreak caused a 68%

reduction in foliar biomass but only a 58% reduction in canopy N pool size.

The canopy N pool is an important regulator of ecosystem N cycling (Prescott 2002), and

declines in canopy N pool size paralleled declines in live foliar biomass during the current

20

outbreak. Foliar N concentrations for undisturbed stands were similar to values previously

reported for the Yellowstone region (Litton et al. 2004) and elsewhere (Binkley et al. 1995,

Blevins et al. 2005). However, we were surprised to observe that foliar N concentration in

current-year needles remained elevated in the 30-year TSB stands, although N concentration of

composite foliage returned to undisturbed levels. Net N mineralization rates were comparable in

undisturbed and 30-year TSB stands but foliage biomass remained substantially lower; thus, the

availability of N per unit of foliage biomass was likely greater in the 30-year stands.

A flush of understory growth has been observed in many post-beetle conifer forest types,

including lodgepole pine (Stone and Wolfe 1996), Douglas-fir (McMillin and Allen 2003),

Engelmann spruce (Picea engelmanni) (Schmid and Frye 1977), subalpine fir (Abies lasiocarpa)

(McMillin et al. 2003), and ponderosa pine (McCambridge et al. 1982). Reduced competition

and increased light, moisture and nutrients are likely stimulants of this growth. As forbs prefer

NO3- uptake over NH4

+ in deciduous woodland (Falkengren-Grerup et al. 2004) and alpine

meadow (Miller and Bowman 2002) systems, increased forb cover observed in this study may be

related to increased net nitrification. The positive relationship between N mineralization and

Carex geyerii %N also suggests that herbaceous species benefit from increased N availability

during bark beetle outbreak. Foliar %N of Carex geyeri can increase after fertilization in

ponderosa pine forests (VanderSchaaf et al. 2004), and in lodgepole pine C. geyerii %N can

increase following fire-induced changes in soil N availability. Interestingly, the increased N

concentration we observed in C. geyerii is comparable to the increase observed two years after

stand-replacing fire in lodgepole pine forests of the same region (Metzger et al. 2006).

21

Interpretations and comparison to post-fire lodgepole pine forests

Multiple lines of reasoning suggest that increased rates of net N mineralization and net

nitrification during the current outbreak are likely not due to a pulse of litter N associated with

beetle-killed trees. Beetle outbreak did not significantly change the mass or N pool size of total

litter, and neither was correlated with N mineralization rates—a pattern that has been noted

across a wide range of North American forests (Scott and Binkley 1997). Furthermore, fresh

lodgepole pine litter immobilizes N for several years, acting as a net sink of N during this period

rather than a source (Fahey 1983, Remsburg and Turner 2006). Thirdly, increases in net N

mineralization following tree mortality are often limited to mineral soil horizons and not found in

the forest floor (Morehouse et al. 2008, Tan et al. 2008, Thiel and Perakis 2009), which suggests

belowground processes and abiotic changes may be more important than litter N inputs. Labile C

leaching from fresh litter or decaying roots could play a role in stimulating gross N production

and N consumption if microbial communities were C limited, although Giardina et al. (2001)

found lodgepole pine soils of the intermountain West to have relatively large amounts of high

quality soil C, with no relationship between C and N mineralization rates. Alternatively, declines

in belowground C transfer beyond the rhizosphere (Högberg et al. 2010) following beetle-

induced mortality could lead to decreased N demand by microbes and thus greater net N

mineralization rates and extractable N pools. Further study is needed to understand the role of C

availability on N transformations following bark beetle disturbance.

Despite the increases in available soil N associated with the current beetle disturbance,

surviving lodgepole pine did not appear to have an excess of available N. Reduced competition

and increased net N mineralization increased nutrient availability for surviving unattacked trees,

as evidenced by a 20-30% increase in foliar N concentration. However, N concentrations in

22

current year foliage remained < 1.2%, suggesting N may still be limiting (Moore et al. 2004).

When foliar N concentrations are considered along with the lack of elevated soil nitrate pools,

our results also suggest that the risk of nitrogen loss via NO3- leaching following bark beetle

outbreak is low.

The effects of bark beetle disturbance on soil and canopy N were substantially less than

observed following fire in Greater Yellowstone. In two-year post-fire lodgepole pine forests,

NH4+ and NO3

- pools were four times greater than in the red stage of a bark beetle outbreak (20

vs 5 µg NH4+-N g soil-1, 2 vs. 0.5 µg NO3

--N g soil-1), and net N mineralization and net

nitrification rates were twice as great (18 vs. 9 µg N g soil-1 year-1 for mineralization, 16 vs. 8 µg

N g soil-1 year-1 for nitrification) (Turner et al. 2007). Several studies also show that foliar N

concentration of lodgepole pine is higher after fire than after bark beetle disturbance. Romme et

al. (2009) found very high (1.87%) foliar N in 3 to 5-year old post-fire lodgepole pine seedlings,

and Turner et al. (2009) found current-year foliage averaged 1.38% N and composite foliage

1.08% N in 17-year old post-fire lodgepole pine. Contrary to fire, bark beetles typically affect

large mature trees with no direct disturbance to the understory or soils. Observed changes in soil

N following bark beetle outbreak are similar to those following other selective agents of

mortality in forests. In eastern hemlock (Tsuga canadensis) forests affected by the hemlock

wooly adlegid (Adelges tsugae Annand), Orwig et al.(2008) documented increased

mineralization, nitrification, and N pool sizes relative to undisturbed stands. Generally, the

consequences of bark beetle outbreak for ecosystem processes are likely similar to any agent of

mortality that selectively kills canopy trees and does not disturb soils or understory directly.

Though our chronosequence approach is strengthened by replication and by validation

from companion dendrochronological analyses (Simard 2011), there were some noteworthy

23

differences among classes. Current outbreak sites had greater soil pH, Ca, and Mg which may

have contributed directly to greater understory cover of forbs. The influence of soil cations on

soil N transformations, however, is considered to be indirect through control on species

composition and subsequent litter quality (Page and Mitchel 2008). Since all sites were

dominated by lodgepole pine, differences in soil cations were unlikely to influence soil N

parameters. Though not significant, thirty year old outbreak sites tended toward greater soil

organic matter and total soil N, which may have resulted from organic N inputs during the

outbreak rather than inherent pre-disturbance site differences.

CONCLUSIONS

Bark beetle disturbance significantly altered N cycling through the litter, soil, and

vegetation of lodgepole pine forests. Beetle-induced litterfall increased litter depth and altered

soil microclimate. Increased soil N in recently disturbed sites was more closely related to cooler

soil temperatures than to litter quality, and was positively correlated with increased foliar N in

unattacked vegetation. Effects of bark beetle disturbance on needle litter %N, soil temperature,

and canopy N pool size persisted 30 years following outbreak. However, changes in soil N

cycling during the outbreak were of lesser magnitude than those observed following stand

replacing fire. Although significant, the net effects of bark beetle disturbance on N cycling in

lodgepole pine were surprisingly minor given the extent of beetle-caused tree mortality.

ACKNOWLDEGEMNTS

We thank Bill Romme, Gary Lovett, Emily Stanley, and Erica Smithwick, and two

anonymous reviewers for constructive comments that improved this manuscript. For help in the

24

field and the laboratory, we appreciate the assistance of many University of Wisconsin students:

Jaclyn Entringer, Heather Lumpkin, Lucille Marescot, Erin Mellenthin, Corey Olsen, Ryan

Peasley, Alex Rahmlow, Amanda Rudie, Ben Ruh, and Greg Skupien.

REFERENCES

Binkley, D., Bell, R., Sollins, P., 1992. Comparison of methods for estimating soil-nitrogen transformations in adjacent conifer and alder-conifer forests. Can. J. For. Res. 22, 858- 863. Binkley, D., Smith, F.W., Son, Y., 1995. Nutrient supply and declines in leaf-area and production in lodgepole pine. Can. J. For. Res. 25, 621-628. Blevins, D.P., Prescott, C.E., Allen, H.L., Newsome, T.A., 2005. The effects of nutrition and density on growth, foliage biomass, and growth efficiency of high-density fire-origin lodgepole pine in central British Columbia. Can. J. For. Res. 35, 2851-2859. Boon, S., 2009. Snow ablation energy balance in a dead forest stand. Hydrol. Process. 23, 2600- 2610. Bormann, F.H., Likens, G.E., 1979. Pattern and Process in a Forested Ecosystem: Disturbance, Development, and the Steady State Based on the Hubbard Brook Ecosystem Study. Springer-Verlag, New York. Bouyoucos, G.J., 1962. Hydrometer method improved for making particle and size analysis of soils. Agron. J. 54, 464-465. Brown, J.K., 1978. Weight and density of crowns of Rocky Mountain conifers. USFS Intermountain Forest and Range Experimental Station. Gen. Tech. Rep. INT-16. Busse, M.D., 1994. Downed bole-wood decomposition in lodgepole pine forests of central Oregon. Soil Sci. Soc. Am. J. 58, 221-227. Byers, J.A., 1984. Electronic multiprobe thermometer and multiplexer for recording temperatures of microenvironments in the forest litter habitat of bark beetles (Coleoptera, Scolytidae). Environ. Entomol. 13, 863-867. Chapin, F.S., Matson, P.A., Mooney, H.A., 2002. Principles of Terrestrial Ecosystem Ecology. Springer, New York. Christenson, L.M., Lovett, G.M., Mitchell, M.J., Groffman, P.M., 2002. The fate of nitrogen in gypsy moth frass deposited to an oak forest floor. Oecologia 131, 444-452.

25

Cullings, K.W., New, M.H., Makhija, S., Parker, V.T., 2003. Effects of litter addition on ectomycorrhizal associates of a lodgepole pine (Pinus contorta) stand in Yellowstone National Park. Appl. Environ. Microbiol. 69, 3772-3776. Dirks, R.A., Martner, B.E., 1982. The climate of Yellowstone and Grand Teton National Parks. US Dept. of the Interior, National Park Service, Washington D.C. Dordel, J., Feller, M.C., Simard, S.W., 2008. Effects of mountain pine beetle (Dendroctonus ponderosae Hopkins) infestations on forest stand structure in the southern Canadian Rocky Mountains. For. Ecol. Manage. 255, 3563-3570. Fahey, T.J., 1983. Nutrient dynamics of above-ground detritus in lodgepole pine (Pinus Contorta ssp Latifolia) ecosystems, southeastern Wyoming. Ecol. Monogr. 53, 51-72. Fahey, T.J., Yavitt, J.B., Pearson, J.A., Knight, D.H., 1985. The nitrogen cycle in lodgepole pine forests, southeastern Wyoming. Biogeochemistry. 1, 257-275. Falkengren-Grerup, U., Michelsen, A., Olsson, M.O., Quarmby, C., Sleep, D., 2004. Plant nitrate use in deciduous woodland: the relationship between leaf N, N-15 natural abundance of forbs and soil N mineralisation. Soil Biol. Biochem. 36, 1885-1891. Giardina, C.P., Ryan, M.G., Hubbard, R.M., Binkley, D., 2001. Tree species and soil textural controls on carbon and nitrogen mineralization rates. Soil Sci. Soc. Am. J. 65, 1272-1279. Griffiths, R.P., Gray, A.N., Spies, T.A., 2010. Soil properties in old-growth Douglas-fir forest gaps in the western Cascade Mountains of Oregon. Northwest Sci. 84, 33-45. Hais, M., Kucera, T., 2008. Surface temperature change of spruce forest as a result of bark beetle attack: remote sensing and GIS approach. Eur. J. For. Res. 127, 327-336. Hawkins, B.J., Boukcim, H., Plassard, C., 2008. A comparison of ammonium, nitrate and proton net fluxes along seedling roots of Douglas-fir and lodgepole pine grown and measured with different inorganic nitrogen sources. Plant Cell Environ. 31, 278-287. Högberg, M. N., M. J. I. Briones, S. G. Keel, D. B. Metcalfe, C. Campbell, A. J. Midwood, B. Thornton, V. Hurry, S. Linder, T. Nasholm, and P. Högberg. Quantification of effects of season and nitrogen supply on tree below-ground carbon transfer to ectomycorrhizal fungi and other soil organisms in a boreal pine forest. New Phytol. 187, 485-493. Houle, D., Duchesne, L., Boutin, R., 2009. Effects of a spruce budworm outbreak on element export below the rooting zone: a case study for a balsam fir forest. Ann. For. Sci. 66, 707. Huber, C., 2005. Long lasting nitrate leaching after bark beetle attack in the highlands of the Bavarian Forest National Park. J. Environ. Qual. 34, 1772-1779.

26

Kizlinski, M.L., Orwig, D.A., Cobb, R.C., Foster, D.R., 2002. Direct and indirect ecosystem consequences of an invasive pest on forests dominated by eastern hemlock. J. Biogeogr. 29, 1489-1503. Klutsch, J.G., Negron, J.F., Costello, S.L., Rhoades, C.C., West, D.R., Popp, J., Caissie, R., 2009. Stand characteristics and downed woody debris accumulations associated with a mountain pine beetle (Dendroctonus ponderosae Hopkins) outbreak in Colorado. For. Ecol. Manage. 258, 641-649. Knight, D.H., Yavitt, J.B., Joyce, G.D., 1991. Water and nitrogen outflow from lodgepole pine forest after 2 levels of tree mortality. For. Ecol. Manage 46, 215-225. Kosola, K.R., Dickmann, D.I., Paul, E.A., Parry, D., 2001. Repeated insect defoliation effects on growth, nitrogen acquisition, carbohydrates, and root demography of poplars. Oecologia. 129, 65-74. Laiho, R., Prescott, C.E., 1999. The contribution of coarse woody debris to carbon, nitrogen, and phosphorus cycles in three Rocky Mountain coniferous forests. Can. J. For. Res. 29, 1592-1603. Litton, C.M., Ryan, M.G., Knight, D.H., 2004. Effects of tree density and stand age on carbon allocation patterns in postfire lodgepole pine. Ecol. Appl. 14, 460-475. Lodge, D.J., McDowell, W.H., McSwiney, C.P., 1994. The importance of nutrient pulses in tropical forests. Trends Ecol. Evol. 9, 384-387. Lovett, G.M., Ruesink, A.E., 1995. Carbon and nitrogen mineralization from decomposing gypsy moth frass. Oecologia. 104, 133-138. Lynch, H.J., Renkin, R.A., Crabtree, R.L., Moorcroft, P.R., 2006. The influence of previous mountain pine beetle (Dendroctonus ponderosae) activity on the 1988 Yellowstone fires. Ecosystems. 9, 1318-1327. McCambridge, W.F., Morris, M.J., Edminster, C.B., 1982. Herbage production under ponderosa pine killed by the mountain pine beetle in Colorado. USDA Forest Service Intermountain Research Station Research Paper RM-416. McMillin, J.D., Allen, K.K., 2003. Effects of Douglas-fir beetle (Coleoptera : Scolytidae) infestations on forest overstory and understory conditions in western Wyoming. West. N. Am. Nat. 63, 498-506. McMillin, J.D., Allen, K.K., Long, D.F., Harris, J.L., Negron, J.F., 2003. Effects of western balsam bark beetle on spruce-fir forests of north-central Wyoming. West. J. Appl. For. 18, 259-266.

27

Metzger, K.L., Romme, W.H., Turner, M.G., 2006. Foliar nitrogen patterns following stand- replacing fire in lodgepole pine (Pinus contorta var. latifolia) forests of the Rocky Mountains, USA. For. Ecol. Manage. 227, 22-30. Metzger, K.L., Smithwick, E.A.H., Tinker, D.B., Romme, W.H., Balser, T.C., Turner, M.G., 2008. Influence of coarse wood and pine saplings on nitrogen mineralization and microbial communities in young post-fire Pinus contorta. For. Ecol. Manage. 256, 59-67. Miller, A.E., Bowman, W.D., 2002. Variation in nitrogen-15 natural abundance and nitrogen uptake traits among co-occurring alpine species: do species partition by nitrogen form? Oecologia 130, 609-616. Moore, J.A., Mika, P.G., Shaw, T.M., Garrison-Johnston, M.I., 2004. Foliar nutrient characteristics of four conifer species in the interior Northwest United States. West. J. Appl. For. 19, 13-24. Morehouse, K., Johns, T., Kaye, J., Kaye, A., 2008. Carbon and nitrogen cycling immediately following bark beetle outbreaks in southwestern ponderosa pine forests. For. Ecol. Manage. 255, 2698-2708. Nigh, G.D., Antos, J.A., Parish, R., 2008. Density and distribution of advance regeneration in mountain pine beetle killed lodgepole pine stands of the montane spruce zone of southern British Columbia. Can. J. For. Res. 38, 2826-2836. Orwig, D.A., Cobb, R.C., D'Amato, A.W., Kizlinski, M.L., Foster, D.R., 2008. Multi-year ecosystem response to hemlock woolly adelgid infestation in southern New England forests. Can. J. For. Res. 38, 834-843. Ostertag, R., Scatena, F.N., Silver, W.L., 2003. Forest floor decomposition following hurricane litter inputs in several Puerto Rican forests. Ecosystems 6, 261-273. Page, B. D., and M. J. Mitchell. 2008. Influences of a calcium gradient on soil inorganic nitrogen in the Adirondack Mountains, New York. Ecol. Appl. 18:1604-1614. Parsons, W.F.J., Knight, D.H., Miller, S.L., 1994. Root gap dynamics in lodgepole pine forest: nitrogen transformations in gaps of different size. Ecol. Appl. 4, 354-362. Pickett, S.T.A., White, P.S. (eds.), 1985. The ecology of natural disturbance and patch dynamics. Academic Press, Orlando, FL. Prescott, C.E., 2002. The influence of the forest canopy on nutrient cycling. Tree Physiol. 22, 1193-1200. Prescott, C. E., G. D. Hope, and L. L. Blevins. 2003. Effect of gap size on litter decomposition and soil nitrate concentrations in a high-elevation spruce-fir forest. Can. J. For. Res. 33, 2210-2220.

28

Raffa, K.F., Aukema, B.H., Bentz, B.J., Carroll, A.L., Hicke, J.A., Turner, M.G., Romme, W.H., 2008. Cross-scale drivers of natural disturbances prone to anthropogenic amplification: the dynamics of bark beetle eruptions. Bioscience 58, 501-517. Remsburg, A.J., Turner, M.G., 2006. Amount, position, and age of coarse wood influence litter decomposition in postfire Pinus contorta stands. Can. J. For. Res. 36, 2112-2123. Romme, W.H., Knight, D.H., Yavitt, J.B., 1986. Mountain pine beetle outbreaks in the Rocky Mountains: regulators of primary productivity. Am. Nat. 127, 484-494. Romme, W.H., Tinker, D.B., Stakes, G.K., Turner, M.G., 2009. Does inorganic nitrogen limit plant growth 3-5 years after fire in a Wyoming, USA, lodgepole pine forest? For. Ecol. Manage. 257, 829-835. Safranyik, L., Carroll, A.L., 2006. The biology and epidemiology of the mountain pine beetle in lodgepole pine forests. In: Safranyik, L., Wilson, B. (eds.), The mountain pine beetle: a synthesis of biology, management, and impacts on lodgepole pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, BC, pp. 3-66. SAS Institute Inc., 2003. SAS/STAT Statistical Software, v. 9.1. SAS Institute, Inc., Cary, NC. Schmid, J. M., and R. H. Frye. 1977. Spruce beetle in the Rockies. USFS General Technical Report RM-49. Scott, N.A., Binkley, D., 1997. Foliage litter quality and annual net N mineralization: comparison across North American forest sites. Oecologia. 111, 151-159. Shore, T.L., Safranyik, L., Hawkes, B.C., Taylor, S.W., 2006. Effects of the mountain pine beetle on lodgepole pine stand structure and dynamics. In: Safranyik, L., Wilson, B. (eds.), The mountain pine beetle: a synthesis of biology, management, and impacts on lodgepole pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, B.C., pp. 95-114. Simard, M., Romme, W.H., Griffin, J.M., Turner, M.G., 2011. Do mountain pine beetle outbreaks change the probability of active crown fire in lodgepole pine forests? Ecol. Monog. 81(1). 3-24. Smithwick, E.A.H., Turner, M.G., Metzger, K.L., Balser, T.C., 2005. Variation in NH4

+ mineralization and microbial communities with stand age in lodgepole pine (Pinus contorta) forests, Yellowstone National Park (USA). Soil Biol. Biochem. 37, 1546-1559. Smithwick, E. A. H., D. M. Kashian, M. G. Ryan, and M. G. Turner. 2009. Long-term nitrogen storage and soil nitrogen availability in post-fire lodgepole pine ecosystems. Ecosystems 12, 792-806.

29

Stadler, B., Muller, T., Orwig, D., 2006. The ecology of energy and nutrient fluxes in hemlock forests invaded by hemlock woolly adelgid. Ecology. 87, 1792-1804. Stone, W.E., Wolfe, M.L., 1996. Response of understory vegetation to variable tree mortality following a mountain pine beetle epidemic in lodgepole pine stands in northern Utah. Vegetatio. 122, 1-12. Stottlemyer, R., Troendle, C.A., 1999. Effect of subalpine canopy removal on snowpack, soil solution, and nutrient export, Fraser Experimental Forest, CO. Hydrol. Process. 13, 2287- 2299. Tan, X., Chang, S.X., Comeau, P.G., Wang, Y.H., 2008. Thinning effects on microbial biomass, N mineralization, and tree growth in a mid-rotation fire-origin lodgepole pine stand in the lower foothills of Alberta, Canada. Forest. Sci. 54, 465-474. Thiel, A.L., Perakis, S.S., 2009. Nitrogen dynamics across silvicultural canopy gaps in young forests of western Oregon. For. Ecol. Manage. 258, 273-287. Titus, B. D., C. E. Prescott, D. G. Maynard, A. K. Mitchell, R. L. Bradley, M. C. Feller, W. J. B. Beese, B. A. Seely, R. A. Benton, J. P. Senyk, B. J. Hawkins, and R. Koppenaal. 2006. Post-harvest nitrogen cycling in clearcut and alternative silvicultural systems in a montane forest in coastal British Columbia. For. Chron. 82, 844-859. Turner, M.G., Smithwick, E.A.H., Metzger, K.L., Tinker, D.B., Romme, W.H., 2007. Inorganic nitrogen availability following severe stand-replacing fire in the Greater Yellowstone Ecosystem. Proc. Natl. Acad. Sci. U.S.A. 104, 4782-4789. Turner, M.G., Smithwick, E.A.H., Tinker, D.B., Romme, W.H., 2009. Variation in foliar nitrogen and aboveground net primary production in young postfire lodgepole pine. Can. J. For. Res. 39, 1024-1035. Turner, M.G., Tinker, D.B., Romme, W.H., Kashian, D.M., Litton, C.M., 2004. Landscape patterns of sapling density, leaf area, and aboveground net primary production in post-fire lodgepole pine forests, Yellowstone National Park (USA). Ecosystems. 7, 751-775. USFS, 2006. Forest Insect and Disease Survey, regions 1, 2, & 4. USDA Forest Service. UWSPAL, 2010. Soil and Plant Analysis. In: Peters, J. (ed.), Wisconsin Procedures for Soil Testing, Plant Analysis and Feed & Forage Analysis. http://uwlab.soils.wisc.edu/procedures.htm. University of Wisconsin Soil and Plant Analysis Lab, Madison, WI. VanderSchaaf, C.L., Moore, J.A., Kingery, J.L., 2004. The effect of multi-nutrient fertilization on understory vegetation nutrient concentrations in inland Northwest conifer stands. For. Ecol. Manage. 190, 201-218.

30

Vitousek, P.M., Gosz, J.R., Grier, C.C., Melillo, J.M., Reiners, W.A., Todd, R.L., 1979. Nitrate losses from disturbed ecosystems. Science 204, 469-474. Vyse, A., Ferguson, C., Huggard, D.J., Roach, J., Zimonick, B., 2009. Regeneration beneath lodgepole pine dominated stands attacked or threatened by the mountain pine beetle in the south central Interior, British Columbia. For. Ecol. Manage. 258, S36-S43. WRCC. 2010. Lake Yellowstone, WY, climate summary. http://www.wrcc.dri.edu/cgi- bin/cliMAIN.pl?wy5345. Western Regional Climate Center.

31

TABLES

Table 1. Site characteristics across a chronosequence of mountain pine beetle disturbance

in P. contorta forests. Superscript letters next to values denote significant differences among

classes (Tukey’s test, α = 0.05). N = 5 per class. Error ranges = 2 SE.

w ANOVA P value; n.s. = non significant x Post-fire stand age at time of mountain pine beetle disturbance. y Data for 30 year TSB sites are from Simard (2011). z Values are an index of southwest-ness ranging from -1 to 1, calculated as: cosine(aspect-225)

TSB Class Site variable Undisturbed Red Gray 30 years P w

Stand agex (year) 179 ± 46 a 164 ± 32 a 176 ± 43 a 152 ± 28 a n.s.

P. contorta basal area (m2ha-1)

Pre-outbreak live y 42.7 ± 10.0 a 48.3 ± 11.9 a 40.2 ± 7.7 a 42.4 ± 9.4 a n.s.

Beetle-killed ---- 41.7 ± 10.5 a 26.6 ± 6.5 a 33.0 ± 12.7 a n.s

Dead, non-beetle killed 5.6 ± 2.4 a 6.0 ± 2.9 a 10.9 ± 3.0 a 6.5 ± 7.2 a n.s.

Topography

Elevation (m) 2400 ± 17 a 2487 ± 34 a 2476 ± 48 a 2218 ± 160 b 0.001

Slope (º) 8 ± 7 ab 10 ± 9 ab 21 ± 11a 2 ± 2 b 0.025

Aspect z 0.10 ± 0.94 a 0.30 ± 0.66 a 0.56 ± 0.26a 0.04 ± 0.81 a n.s.

General Soil Characteristics

Sand (%) 62 ± 6 ab 58 ± 5 ab 69 ± 7 a 55 ± 6 b 0.034

Silt (%) 26 ± 6 a 27 ± 5 a 20 ± 6 a 31 ± 5 a n.s.

Clay (%) 13 ± 1 a 15 ± 2 a 11 ± 2 a 13 ± 2 a n.s.

Bulk density (g cm-3) 0.7 ± 0.1ab 0.7 ± 0.2 ab 0.9 ± 0.2 a 0.6 ± 0.1 b 0.042

pH 4.9 ± 0.1 b 5.4 ± 0.1 a 5.2 ± 0.1 a 4.9 ± 0.2 b <0.001

Ca (ppm; exchangeable) 452 ± 63 b 1192 ± 651 a 857 ± 244 ab 322 ± 182 b 0.014

Mg (ppm;exchangeable) 67 ± 12 b 143 ± 54 a 117 ± 35 ab 71 ± 21 b 0.018

K (ppm; exchangeable) 162 ± 22 a 186 ± 60 a 149 ± 29 a 145 ± 23 a n.s.

P (ppm; Bray P1) 17 ± 4 a 25 ± 12 a 20 ± 9 a 11 ± 3 a n.s.

OM (%) 3.9 ± 0.8 a 5.3 ± 3.9 a 3.3 ± 0.4 a 5.3 ± 1.4 a n.s.

C:N 72 ± 23 a 42 ± 26 a 41 ± 7 a 42 ± 11 a n.s.

Total N (%) 0.04 ± 0.03 a 0.05 ± 0.03 a 0.04 ± 0.01a 0.08 ± 0.04 a n.s.

Organic N (µg N g soil-1) 368 ± 271 a 509 ± 306 a 446 ± 109 a 759 ± 416 a n.s.

32

Table 2. Air and soil temperatures across a chronosequence of mountain pine beetle

disturbance. Superscript letters next to values denote significant differences among classes

(Tukey’s test, α = 0.05). N = 3 per class. Error ranges = 2SE.

w ANOVA P value; n.s. = non significant x Values are an index of southwest-ness ranging from -1 to 1, calculated as: cosine (aspectº-225)

TSB Class

Variable Undisturbed Red Gray 30 years P w

Air temperature (ºC)

Mean daily temperature 15.3 ± 0.4 b 15.1 ± 0.3 b 15.0 ± 0.5 b 16.3 ± 0.5 a 0.006

Mean daily range 19.4 ± 1.4 b 21.8 ± 0.8 ab 20.5 ± 1.6 b 25.1 ± 2.8a 0.012

Litter-soil temperature (ºC)

Mean daily temperature 15.4 ± 1.0 b 13.3 ± 0.8 bc 13.0 ± 0.6 c 19.2 ± 1.4 a < 0.001

Mean daily range 27.4 ± 3.5 b 21.8 ± 3.6 b 19.3 ± 1.0 b 38.0 ± 7.0 a 0.001

Mean daily difference from air 0.1 ± 0.8 b -1.8 ± 1.0 b -2.0 ± 0.2 b 2.8 ± 1.4 a <0.001

10 cm soil depth temperature (ºC)

Mean daily temperature 10.5 ± 0.3 b 9.4 ± 0.1 b 9.7± 0.4 b 13.8 ± 1.3 a < 0.001

Mean daily range 2.4 ± 0.2 a 2.9 ± 0.8 a 2.8 ± 0.6 a 4.6 ± 1.8 a n.s.

Mean daily difference from air -4.8 ± 0.1 b -5.6 ± 0.3 b -5.3 ± 0.8 b -2.5 ± 0.9 a < 0.001

Togography of instrumented sites

Elevation (m) 2401 ± 4 ab 2471 ± 40 a 2494 ± 68 a 2198 ± 227 b 0.0301

Slope (º) 10.8 ± 10.6 a 6.2 ± 3.4 a 13.7 ± 5.5 a 2.5 ± 3.2 a n.s.

Aspect x 0.26 ± 1.25 a 0.42 ± 0.70 a 0.55 ± 1.34a -0.09 ± 0.67 a n.s.

33

Table 3. General linear models of net N mineralization and net nitrification across a

chronosequence of mountain pine beetle disturbance in P. contorta forests. N = 5 per class.

Response variable Model Fit (R2, F, P) Parameter Estimate F P

Net N mineralization 0.70, 5.04, 0.007 TSB class -- 5.30 0.01

Undisturbed -- -- --

Red 18.2 -- 0.02

Gray 16.2 -- 0.03

30 years -8.6 -- 0.09

Bulk density -32.02 7.68 0.02

pH -22.17 4.37 0.06

C:N -0.19 5.66 0.03

Intercept 153.9 -- 0.02

Net nitrification 0.52, 4.06, 0.02 TSB class -- 4.58 0.02

Undisturbed -- -- --

Red 1.3 -- 0.01

Gray 1.2 -- 0.04

30 years -0.31 -- 0.53

Bulk density -3.56 8.52 0.01

Intercept 4.01 -- 0.0005

34

Table 4. Foliar N in unattacked lodgepole pine (P. contorta) and an understory sedge (C.

geyerii) across a chronosequence of mountain pine beetle disturbance in P. contorta forests.

Superscript letters denote significant differences among classes (Tukey’s HSD test, α = 0.05).

Error ranges = 2SE.

TSB Class

Foliage variable Undisturbed Red Gray 30 years P w

Pinus contorta

New foliage

N (%) 0.82 ± 0.03 b 1.05 ± 0.12 a 1.00 ± 0.06 a 0.94 ± 0.05 a 0.016

C:N 59 ± 4 a 49 ± 6 b 50 ± 2 b 53 ± 2 ab 0.022

All foliage

N (%) 0.75 ± 0.04 b 0.91 ± 0.08 a 0.91 ± 0.09 a 0.75 ± 0.12 b 0.014

C:N 63 ± 3 a 52 ± 3 b 53 ± 5 b 64 ± 10 ab 0.017

Biomass

(g m-2) 1107 ± 214 a 331 ± 88 b 358 ± 169 b 571 ± 78 b <0.0001

N pool size

(g N m-2) 8.3 ± 1.7 a 3.0 ± 0.9 b 3.5 ± 1.5 b 4.3 ± 0.8 b <0.0001

Carex geyerii

N (%) 1.1 ± 0.2 b 1.6 ± 0.4 a 1.5 ± 0.1 ab 1.1 ± 0.1 b 0.016

C:N 43 ± 9 a 28 ± 7 b 29 ± 2 ab 41 ± 4 ab 0.024

Biomass

(g m-2) 46 ± 46 a 32 ± 42 a 52 ± 42 a 179 ± 129 a n.s.

N pool size

(g N m-2) 0.5 ± 0.5 a 0.5 ± 0.8 a 0.6 ± 0.8 a 2.1 ± 1.7 a n.s.

35

FIGURE LEGENDS

Figure 1. Location of chronosequence sites in the Greater Yellowstone region. A) All sites; B)

Undisturbed and 30 yr TSB (time-since-beetle) sites in Yellowstone National Park; C) Red and

Gray sites in the Green River Lakes region of Bridger-Teton National Forest.

Figure 2. Understory cover by plant functional group across a chronosequence of mountain pine

beetle disturbance in lodgepole pine forest.

Figure 3. Litter variables across a chronosequence of mountain pine beetle disturbance in

lodgepole pine. A) litter depth, B) litter mass, C) fresh needle litter %N, D) all-needle litter %N,

E) total litter %N, F) litter N pool. P values are from ANOVA, and class differences were

determined using Tukey’s HSD test (α = 0.05). N = 5/class, error bars = 2 standard errors.

Values within the same panel with the same letter are not significantly different.

Figure 4. Soil N pools and fluxes across a chronosequence of mountain pine beetle

disturbance in lodgepole pine. For extractable NH4+ pool size (panel A), and extractable NO3

-

pool size (panel B), significance levels are from ANOVA and Tukey’s HSD test (α = 0.05). For

net mineralization rate (panel C), and net nitrification rate (panel D), significance levels were

determined from general linear models reported in Table 3. Summary statistics in each panel are

for the class terms in each respective mode, and values within the same panel with the same

letter are not significantly different. Error bars = 2 standard errors.

36

Figure 5. Relationships between litter depth and N content, soil temperatures, and net N

mineralization rates across a chronosequence of mountain pine beetle disturbance in

lodgepole pine. ● = undisturbed, ■ = red stage, ▲ = gray stage, ♦ = 30 year after outbreak. A)

Soil temperature vs. litter depth. Solid symbols = litter-soil interface temperatures, open symbols

= 10-cm soil depth temperatures; B) Net N mineralization vs. litter-soil interface temperature; C-

E) Net N mineralization vs. fresh needle litter %N (C), all needle litter %N (D), and total litter

%N (E); F) Net N mineralization vs. litter N pool. In panels A and B, n = 12 as only three plots

per class were instrumented with temperature probes; All plots are included in panels C-F (n =

20). Summary statistics and regression lines in each panel are for significant linear regressions.

Figure 6. Foliar %N of unattacked lodgepole pine (A) and %N of the sedge Carex geyerii (B)

vs. soil N mineralization rate in a chronosequence of mountain pine beetle disturbance in

lodgepole pine. ● = undisturbed, ■ = red stage, ▲ = gray stage, ♦ = 30 year after outbreak.

Summary statistics are from linear regression.

37

FIGURES

Figure 1

38

Figure 2

0

10

20

30

40

50

60

70

Undisturbed Red Gray 30 yr

% g

roun

d co

ver

SeedlingsForbShrubSedgeGrass

39

Figure 3

0.00.51.01.52.02.53.03.54.0

Undisturbed Red Gray 30 yrs

Litte

r dep

th (c

m)

BC

A

C

ABP =0.002

0

500

1000

1500

2000

2500

3000


Litte

r mas

s (g

m-2

)

P =n.s

0

5

10

15

20

25


Litte

r N p

ool (

g N

m-2

)

AB AB

B

AP =0.04

A)

B)

C)

0.0

0.2

0.4

0.6

0.8

1.0


Fres

h ne

edle

litte

r %N

B

A

B

AP < 0.0001

0.0

0.2

0.4

0.6

0.8

1.0


All

need

le li

tter %

N

P < 0.0001

CB

ABA

0.0

0.2

0.4

0.6

0.8

1.0


Tota

l litt

er %

N

P =n.s.

D)

E)

F)

40

Figure 4

0

5

10

15

20

25

30


Net

nitr

ifica

tion

(µg

N g

soi

l-1 y

r-1)

P = 0.02D)

0

5

10

15

20

25

30


Net

N m

iner

aliz

atio

n (µ

g N

g s

oil-1 yr

-1)

P = 0.01C)

0

1

2

3

4

5

6

7

8


Ext

ract

able

NH4

+ (µg

N g

soi

l-1)

P = 0.05A)

B

ABAB

A

C

AB

BC

A

B

A

A

B

0

1

2

3

4

5

6

7

8

Undisturbed Red Gray 30 yrE

xtra

ctab

le N

O3- (µ

g N

g s

oil-1

)

P = n.s.B)

41

Figure 5

6

8

10

12

14

16

18

20

22

0.5 1.0 1.5 2.0 2.5 3.0Litter depth (cm)

Tem

pera

ture

(C)

Adj. R 2 = 0.31; P = 0.04

Adj. R 2 = 0.30; P = -226

10141822263034

12 14 16 18 20 22Temperature (C)

Net

N m

iner

aliz

atio

n(µ

g N

g s

oil-1

yr-1

)

Adj. R 2 = 0.39;P = 0.02

-226

10141822263034

0.2 0.4 0.6 0.8 1.0All needle litter %N

Net

N m

iner

aliz

atio

n(µ

g N

g s

oil-1

yr-1

)

P = n.s.

-226

10141822263034

0.2 0.4 0.6 0.8 1.0Fresh needle litter %N

Net

N m

iner

aliz

atio

n(µ

g N

g s

oil-1

yr-

1)

Adj. R 2 = 0.19; P = 0.03

-226

10141822263034

0.2 0.4 0.6 0.8 1.0Total Litter %N

Net

N m

iner

aliz

atio

n(µ

g N

g s

oil-1

yr-1

)

P = n.s.

-226

10141822263034

0 5 10 15 20 25 30Total litter N pool (g litter N m-2)

Net

N m

iner

aliz

atio

n(µ

g N

g s

oil-1

yr-1

)

P = n.s.

A)

D)

B)

C)

F)E)

42

Figure 6

0.75

0.85

0.95

1.05

1.15

1.25

P. c

onto

rta f

resh

folia

r % N

A) Adj. R 2 = 0.41; P = 0.003

0.75

1.00

1.25

1.50

1.75

2.00

-5 0 5 10 15 20 25 30 35

Net N mineralization (µg N g soil-1 yr-1)

C. g

eyer

i %

N

B) Adj. R 2 = 0.48 ;P = 0.007

43

CHAPTER 2

Forest type influences ecosystem response to bark beetle disturbance

ABSTRACT

Outbreaks of native Dendroctonus bark beetles are causing extensive tree mortality in

conifer forests throughout western North America, yet consequences for nitrogen (N) cycling in

different forest types are not well known. We quantified beetle-induced changes in forest

structure, soil microclimate, and N cycling through litter, soils, and vegetation in Douglas-fir

(Pseudotsuga menziesii) and lodgepole pine (Pinus contorta var. latifolia) forests of Greater

Yellowstone (WY, USA). We hypothesized N cycling responses would be greater in Douglas-fir,

which generally has higher overall N capital. Undisturbed Douglas-fir stands had larger litter and

soil N pools, and greater net N mineralization rates than lodgepole pine. However, responses to

disturbance were similar and included a pulse of N-enriched litter, a doubling of soil N

availability, a 30-50% increase in understory cover, and a 20% increase in N concentration of

composite foliage in unattacked trees. Soil temperature did not vary in Douglas-fir, but was

cooler in beetle-disturbed lodgepole pine. N concentration in fresh foliage was uncorrelated with

N mineralization in Douglas-fir, but positively correlated in lodgepole pine. Though pools and

fluxes of soil inorganic N doubled, they remained low in both forest types (< 6 µg N g soil-1

NH4+-N or NO3

--N; < 25 µg N g soil-1 yr-1 net N mineralization; < 8 µg N g soil-1 yr-1 net

nitrification). Our results suggest that bark beetle disturbance affects N cycling similarly in

Douglas-fir and lodgepole pine, and both systems appear to have mechanisms that could enhance

N retention following beetle outbreak.

44

INTRODUCTION

Insect outbreaks contribute significantly to many disturbance regimes, and can be strong

drivers of change in forest ecosystems (Haack and Byler 1993). Insect-induced mortality alters

ecosystem structure by redistributing biomass among ecosystem pools, often via the selective

mortality of mature canopy trees. In turn, vegetation-driven aspects of ecosystem function such

as primary production and elemental cycling may be affected from local to regional scales

(Schowalter 1981). Widespread outbreaks of native Dendroctonus bark beetles are occurring

within multiple conifer forest types of western North America, yet the consequences of this

disturbance type for many ecosystem functions are not well known. Furthermore, the pre-

outbreak condition and biogeochemical characteristics of infested stands can vary substantially

among forests dominated by different host tree species (Thomas and Prescott 2000, Giardina et

al. 2001, Lovett et al. 2004). In turn, ecosystems organized around such foundational species

(Ellison et al. 2005) could have unique responses to similar bark beetle disturbance (Lovett et al.

2006). This variation may arise from specific biogeochemical properties of the host (e.g. nutrient

uptake rates; nutrient allocation among tissues), host landscape position (e.g. elevation, slope,

soil type), or post-disturbance successional trajectory (Schowalter 1981).

Extensive outbreaks of the Douglas-fir beetle (DFB; Dendroctonus pseudotsugae

Hopkins) and mountain pine beetle (MPB; Dendroctonus ponderosae Hopkins) have affected

Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), and lodgepole pine (Pinus contorta Dougl.)

forests of Greater Yellowstone (Wyoming, USA) since 2003. At epidemic population levels,

these closely related Dendroctonus species utilize pheromone-mediated mass attack strategies to

overcome host trees (Raffa and Berryman 1987) and produce similar patterns of tree mortality,

litterfall, and canopy decay across host forest types. Both beetle species tend to attack large

45

canopy-dominant trees. Stand-level mortality rates frequently exceed 50% (Negron 1998,

Safranyik and Wilson 2006), leaving post-outbreak stands with decreased live stem density and

mean stem diameter (McMillin and Allen 2003, Shore et al. 2006). Foliage of beetle-killed trees

of both species turns red in the year following successful attack (red stage), and is shed to the

forest floor within two or three years (gray stage) (Negron 1998, McMillin and Allen 2003).

Canopy opening and subsequent inputs of needle litter to the forest floor induced by bark beetles

may influence soil microclimate (Griffin et al. 2011) by increasing soil insulation (Byers 1984),

incident radiation (Hais and Kucera 2008), and air flow through the stand (Boon 2009).

Despite the similar mortality, litterfall, and structural changes induced by both the MPB

and DFB in their respective host forest types, subsequent effects on nitrogen (N) cycling may

differ. Although lodgepole pine litter has a slightly higher N content than Douglas-fir, the

concentration of lignin in lodgepole pine litter is twice that of Douglas-fir, resulting in a greater

lignin:N ratio (Thomas and Prescott 2000). Lignin:N ratio is a strong control on litter

decomposition (Scott and Binkley 1997) and subsequent return of N to the soil profile (Prescott

2005). Higher decay constant (k) values of Douglas-fir litter relative to lodgepole pine support

this relationship (Keane 2008). Thus, in undisturbed systems N may be retained in the litter layer

longer in lodgepole pine forests than in Douglas-fir. Insect-induced changes in needle senescence

and nutrient content are known to influence litter decomposition rates in conifers (Koukol et al.

2008, Przybyl et al. 2008), and bark beetle mortality has been shown to increase fresh needle

litter N concentration in both ponderosa pine (Pinus ponderosae Laws.) and lodgepole pine

(Morehouse et al. 2008, Griffin et al. 2011). Furthermore, in undisturbed forests Douglas-fir

generally has greater pool sizes of soil inorganic N and greater rates of net N mineralization

relative to lodgepole pine (Thomas and Prescott 2000). Greater baseline N capital of soils is

46

correlated with accelerated N cycling between soil, foliage, and litter N pools (Prescott et al.

2000), and may increase the potential for nutrient loss following disturbance. Laboratory studies

show uptake rates of inorganic N in Douglas-fir roots are 73% greater for NO3- and 35% greater

for NH4+ than in lodgepole pine (Hawkins et al. 2008). These differences suggest that declines in

stand-level nutrient uptake following bark beetle outbreak may be greater in Douglas-fir forests

relative to lodgepole pine, leading to greater accumulation of inorganic N in soils and thus a

greater potential for N leaching. Differences in N uptake rates may also contribute to variable

responses of foliar N content in unattacked or surviving trees following disturbance.

In this study we address two overarching questions: (1) How does DFB disturbance alter

forest structure, soil microclimate, and N cycling through litter, soils, and vegetation in

Douglas-fir forests? (2) Are the patterns of change observed in beetle-disturbed Douglas-fir

forests similar to those in mountain pine beetle-disturbed lodgepole pine forests? Forest

structure, soil microclimate, and nitrogen pools and fluxes were measured in both undisturbed

and gray stage beetle-killed forests. Hypothesized bark beetle effects and suggested mechanisms

are summarized in Figure 1. We expected beetle disturbance to decrease live canopy biomass

and initiate a pulse of N-enriched litter to the forest floor. Shading by litter and growth of

understory vegetation was expected to cool soil temperatures, while soil N cycling rates and

pools of inorganic N were predicted to increase resulting in increased foliar %N of surviving

unattacked trees. When comparing forest types, we expected the absolute magnitude of increases

in soil N pools and fluxes to be larger in Douglas-fir. However, proportional increases were

predicted to be larger in lodgepole pine and therefore potentially more ecologically significant.

Foliar N response to changing soil N availability could also vary among forest type, with more

N-limited stands showing larger increases in foliar N. Bark beetle disturbance may also influence

47

the cycling of micronutrients in conifer forests, which can be important controls on

macronutrient cycles.

METHODS

Lodgepole pine study area, site characteristics, and data used for our comparisons of

forest type were reported in Griffin et al. (2011), with the exception of foliar micronutrients.

Vegetation, litter, and soil sampling methods in Douglas-fir were identical to those used by

Griffin et al. (2011) in mountain pine beetle-disturbed lodgepole pine stands.

Douglas-fir study area and sampling design

Douglas-fir study sites were located in the Greater Yellowstone Ecosystem of

northwestern Wyoming, USA, within Grand Teton National Park and Bridger-Teton National

Forest in the Moran Junction/Buffalo River Valley Region. July temperatures average 14.9 ºC,

January temperatures average -11.2 ºC, with 595 mm of annual precipitation occurring mostly as

snow (WRCC 2010c). The most recent outbreak of DFB in this area began in the early 2000’s,

though the region has experienced widespread outbreak of multiple bark beetle species in recent

years including the mountain pine beetle in lodgepole pine and whitebark pine (Pinus albicaulis

Engelmann), and the spruce beetle (Dendroctonus rufipennis Kirby) in Engelmann spruce (Picea

engelmannii Parry) (USFS 2006). In 2008 we sampled ten spatially independent (>2 km apart)

Douglas-fir stands (five undisturbed, five attacked by bark beetles in 2003 and 2004). DFB-

disturbed stands were identified using aerial surveys of insect damage (USFS 2006), and field

inspection assured selection of stands with comparable structure, basal area, soils, and

disturbance intensity. Within each stand, an 8m radius (201 m2) plot was installed with GPS

location, elevation, and slope recorded at the plot center.

48

Vegetation

All live and dead trees greater than breast height (1.4m) were identified to species and

measured for diameter at breast height (DBH). Beetle-killed trees were identified by the presence

of exit holes on the bark exterior and distinctive Dendroctonus galleries underneath the bark. For

the beetle-disturbed sites, pre-outbreak basal area was calculated by summing the basal area of

live and beetle-killed trees. Mortality rates in disturbed plots were as high as 100% of canopy

trees, and survivors were mostly smaller sub-canopy trees contributing little basal area. Thus, the

post-disturbance growth (~4yrs worth) of survivors was not considered in calculating the pre-

disturbance basal area of beetle-killed stands. Canopy biomass was determined for each tree

using allometrics developed for Rocky Mountain conifers (Brown 1978) and summed by plot.

Ground cover was visually estimated to the nearest 10% by plant functional groups (forbs,

sedges, grasses, shrubs, and tree seedlings) in ten 0.25-m2 circular microplots; total cover was

allowed to exceed 100% to account for multiple strata of ground vegetation. The microplots were

located within the inner 5-m plot radius using a stratified random design of fixed distances (one

at 0.5-m; two at 1.5-m, 2.5-m and 3.5-m; and three at 4.5-m) and random bearings (in 10º

increments) from the plot center, and averaged by plot.

Microclimate

To evaluate variation in growing-season soil microclimate, soil temperature was

measured hourly in three of the five plots within each disturbance class (N = 6 instrumented

plots) from Ju1y 4 through August 31, 2008, using three pairs of iButton datalogger probes

(Maxim Integrated Products Inc., Dallas Semiconductor, Sunnyvale, CA) per plot. One iButton

of each pair was installed at the litter-organic soil interface, and the second was installed 10 cm

below this interface. For each soil depth, temperature probe data were summarized as follows.

49

First, plot-level hourly temperatures were determined by averaging data from the three probes

per plot. Plot-level daily mean temperatures, ranges, maximums, and minimums were then

calculated from the hourly data, and averaged to determine disturbance class-level daily means.

Plot-level growing season values were obtained by averaging daily values across the sampling

period, and disturbance class-level growing season means were then calculated from these plot-

level data.








was composited by plot for each category, and ground to powder for C:N analysis on a Leco

CNS-2000 at the University of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL 2010).

Soil chemistry, N pools, and N fluxes

One soil core was collected from each 0.25-m2 microplot (n = 10 per plot) using a 5-cm

diameter x 15-cm long PVC corer. Soils were sieved (2 mm), weighed, and divided into three

subsamples: 30 g oven-dried at 60 ºC for gravimetric percent moisture; 20 g extracted in 75 ml of

2M KCl for 2 hours, with the extract then filtered and frozen for later analysis of NH4+ and NO3

-

pools; and 20 g air-dried and bulked by plot for soil texture and chemical analyses. Air-dried soil

was analyzed for pH, total N, exchangeable Ca, Mg, and K, available P (Bray P1 extract), and

organic matter at the University of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL

50

2010). Soil organic N was determined by difference using total N and inorganic N values, and

soil texture was determined using the Bouyoucos hydrometer technique (Bouyoucos 1962).

Net N mineralization and net nitrification were measured using ion-exchange resin cores

(Binkley et al. 1992), with one core incubated in situ for approximately one year (July 2008 -

June 2009) in each 0.25-m2 microplot. Incubated cores consisted of a 5-cm diameter x 15-cm

long PVC tube of soil with an ion-exchange resin bag placed at the bottom. Resin bags were

constructed using 20 g of mixed bed ion exchange resin (J.T Baker #JT4631-1) tied inside a

piece of un-dyed nylon stocking material. Upon retrieval in summer 2009, core soils were sieved

(2 mm) and core soils and resin bags were extracted separately in 2M KCl in the same manner as

the 2008 initial soil samples described above. All KCl extractions were analyzed for [NH4+] and

[NO3-] using colorimetric methods on an Astoria Pacific II continuous flow autoanalyzer. For

each microplot, mineralization and net nitrification rates were calculated as:

rate = [(final soil N + resin bag N) – initial soil N ] / incubation time

and expressed in units of µg N g soil-1 year-1. Values were then averaged by plot. Atmospheric N

inputs in the region are low (~ 1 kg N ha-1 year-1; www.epa.gov/castnet/charts/YEL408totn.png)

and were not factored into the soil N calculations.

Foliar Chemistry

In each plot, pole pruners were used to collect canopy foliage from three live trees

unattacked by bark beetles. Two fully sunlit branches (0.5-m long) were clipped from each tree

and separated into two subsamples: current-year foliage (from one branch per tree), and all-years

foliage (from the second branch per tree). Foliar samples were oven-dried at 60ºC and ground to

powder for chemical analyses at the University of Wisconsin Soil and Plant Analysis Lab. C and

N analysis for both current-year fresh foliage and all-year composite foliage was performed on a

51

Leco CNS-2000 analyzer. In current year fresh foliage, P, K, Ca, Mg, S, Zn, B, Mn, Fe, Cu, Al,

and Na were determined by ICP-OES (Thermo Jarrell Ash IRIS Advantage Inductively Coupled

Plasma Optical Emission Spectrometry) and ICPMS (VG Plasma Quad PQ2 Turbo Plus

Inductively Coupled Plasma Mass Spectrometry) (UWSPAL 2010). Tree-level canopy data (N=3

per sample type per plot) were averaged to obtain plot-level data. To compute canopy N pool

sizes, the mean foliar N concentration for each plot was multiplied by biomass values derived

from the allometrics cited above.


All statistical analyses were performed in SAS (SAS Institute Inc. 2003), and unless

otherwise noted all reported variance values are two standard errors. All variables were checked

for normality and transformed if necessary to satisfy the assumptions of statistical methods. N =

10 (5 per disturbance class) for all analyses of Douglas-fir forests except those using

microclimate data, where N = 6 as only three of five plots per disturbance class were

instrumented. For all analyses comparing forest types, N = 20.

General linear models were used to test for differences between both undisturbed and

beetle-killed Douglas-fir forests, as well as undisturbed Douglas-fir and undisturbed lodgepole

pine, in basal areas, site characteristics (slope, elevation, aspect, and soil chemistry), understory

cover, soil temperatures, litter quantity and quality, soil N pools, foliar N concentrations, and

foliar N pools. We also used general linear models to account for potential covariate effects of

soil pH, bulk density, and C:N ratio in addition to disturbance class when testing for differences

in net mineralization and net nitrification rates in Douglas-fir sites. Relationships between soil N

availability and fresh foliar N concentrations were explored using Pearson correlation and linear

regression. Linear regression was also used to test whether abiotic changes in soil temperature or

52

biotic changes in litter N were more closely related to changes in soil N cycling. To test whether

litter N, soil N, and foliar chemistry of lodgepole pine and Douglas-fir forests responded

differently to bark beetle disturbance, we used general linear models of the form: y = species +

disturbance class + species*disturbance class. A significant (P ≤ 0.05) interaction term in these

analyses indicates differing response to beetle disturbance by type.

RESULTS

Douglas-fir forests

Site characteristics, stand structure, and soil microclimate. Undisturbed and beetle-

killed Douglas-fir stands occupied similar landscape positions, with no difference in either slope

or aspect between disturbance classes (Table 1). However, due to the spatial distribution of the

Douglas-fir beetle outbreak, beetle-killed sites were on average 140 m higher in elevation than

undisturbed (2186 vs. 2325 m; P = 0.0400; Table 1). Soil texture (%sand, silt, clay), bulk

density, and organic soil horizon depth were similar between disturbance classes, as were soil

Ca, Mg, K, P, and pH (Table 1). Douglas-fir basal area comprised 97% (67 of 69 m2 ha-1) of total

stand basal area among all sites and did not differ between undisturbed and beetle-killed sites

(Table 1). Live Douglas-fir basal area averaged 98% of the total Douglas-fir basal area in the

undisturbed class and 7% in beetle-killed stands (Table 1). In beetle-killed stands, Douglas-fir

mortality averaged 93% of stem density, with an estimated concurrent decline in live canopy

biomass averaging 15,796 ± 1704 kg ha-1 (Table 1). Total biotic understory cover increased

significantly from 79 to 103%, driven by significant increases in forb (19 vs. 36 %) and grass (18

vs 49 %) cover (P ≤ 0.05; Figure 2). Growing season soil temperatures were not significantly

different between Douglas-fir disturbance classes at either soil depth, averaging 13.1 ± 1.0 ºC at

53

the litter-soil interface and 10.9 ± 0.9º C at 10 cm depth below the litter layer (Table 1).

Variability in soil temperature was also consistent between disturbance classes at both depths,

with no differences in daily temperature means, ranges, minimums, or maximums (Table 1).

Litter quantity and quality. Litter quantity did not differ between undisturbed and beetle-

killed Douglas-fir stands; litter depth averaged 3.2 ± 0.4 cm, and total litter mass averaged 1842

± 247 g m-2 (Table 1). However, litter quality did differ between disturbance classes. The N

concentration in needle litter of all ages increased by 55% in beetle killed stands (from 0.89 to

1.38 %N), and total litter layer N concentration increased by 25% (from 1.10 to 1.37 %N; Table

2). Current year needle litter N concentration showed a marginally significant 13% increase

(from 0.70 to 0.79, P = 0.0641; Table 2). Increases in litter %N resulted in significant concurrent

declines in litter C:N ratio in all three litter categories. However, total litter N pool size (product

of total litter mass and total litter %N) showed only a marginally significant increase (P =

0.0666) from 19.3 to 26.4 g N m-2 (Table 2), as litter mass varied considerably among sites.

Soil N pools and fluxes. Soil inorganic N pools and soil N transformation rates differed

between undisturbed and beetle-killed Douglas-fir stands (Table 2). Extractable ammonium in

beetle-disturbed stands was twice that of undisturbed stands (5.0 ± 0.8 vs. 2.5 ± 0.6 µg N g soil-

1), extractable nitrate was almost four times greater (5.5 ± 5.2 vs. 1.5 ± 1.0 µg N g soil-1), and

total extractable inorganic N (NH4+ + NO3

-) in beetle-disturbed stands was more than double that

in undisturbed (10.5 ± 4.7 vs. 4.0 ± 1.7µg N g soil-1). Annual net N mineralization rates were

highly variable and were not significantly different between disturbance classes, though a

marginally significant increase (P = 0.0611) was found when using soil bulk density as a

covariate, with higher rates observed in beetle-killed stands. Annual net nitrification rates

increased 50% in beetle-disturbed stands relative to undisturbed (7.2 ± 1.0 and 4.8 ± 0.6 µg N g

54

soil-1 yr-1 respectively). Neither net N mineralization nor net nitrification rates were significantly

correlated with any metric of soil temperature (Pearson P < 0.05; N =6), nor was net N

mineralization correlated with any metric of litter N (Pearson P < 0.05; N = 10). Nitrification

fraction (the ratio of net nitrification to net mineralization) averaged 0.38 ± 0.10 and did not

differ among disturbance classes, nor did soil C:N, soil %N, or soil N pool size (Table 2).

Foliar N. Beetle disturbance impacted both foliar N concentration and foliar N pool size.

N concentration in the fresh foliage (current year) of unattacked Douglas-fir trees averaged 1.28

± 0.11% and did not differ between disturbance classes. However, N concentration in all foliage

of unattacked Douglas-fir was 18% greater in beetle-disturbed stands than in undisturbed (0.90 ±

0.07 vs. 0.76 ± 0.04 %N), with a concurrent decrease in C:N ratio (Table 2). Foliar N pool size

mirrored changes in live foliar biomass and declined sharply by 79% from 11.5 ± 1.7 to 2.4 ± 1.5

g N m-2 (Table 2). Neither fresh foliar %N or C:N were significantly correlated with any metric

of inorganic soil N, including NH4+, NO3

-, total inorganic soil N, net mineralization, net

nitrification, or soil C:N (Pearson correlation; data not shown). However, foliar %N of all

needles was positively related to both soil NO3- (linear regression P = 0.0553, F = 5.03, Adj. R2

= 0.31) and total inorganic soil N (linear regression P = 0.0132, F = 10.06, Adj. R2 = 0.50), and

the C:N ratio of all needles was negatively related to total inorganic soil N (linear regression P =

0.0253, F = 7.53, Adj. R2 = 0.42).

Forest type comparison

Landscape position (slope and aspect) did not differ among Douglas-fir and lodgepole

pine sites (P > 0.08) though Douglas-fir sites were approximately 200m lower in elevation (2255

± 71m vs. 2438 ± 35m; P < 0.0001; Douglas-fir data in Table 1, lodgepole pine data reported in

Griffin et al. (2011)), consistent with the species’ distribution in the region (Despain 1990).

55

Douglas-fir soils were higher in pH and contained less sand and more silt than lodgepole pine

soils, as well as greater concentrations of organic matter, Ca, Mg, K, and P (P < 0.01; Douglas-

fir data in Table 1, lodgepole pine data reported in Griffin et al. (2011)). Compared to lodgepole

pine, Douglas-fir sites contained higher total basal area (all sites: 69.0 ± 8.5 vs. 42.7 ± 7.6; P =

0.0003; Figure 2), beetle-killed basal area (beetle-killed sites: 69.6 ± 11.2 vs. 13.7 ± 5.8; P <

0.0001; Figure 2), and percent mortality of pre-outbreak living stems <1.4m DBH (93 ± 3% vs.

52 ± 17%).

Undisturbed forests. In the absence of bark beetles, forest types differed in understory

cover composition and soil temperature. Total biotic cover was greater in Douglas-fir (79 ± 6%

vs. 29 ± 10 %; P < 0.0001), driven by greater cover of shrubs, total graminoids (grasses +

sedges), and tree seedlings compared to lodgepole pine (P < 0.02; Figure 2). Soil temperatures at

the litter-soil interface in undisturbed forests were cooler and had a lower range in Douglas-fir

(temperature: 13.5 ± 0.7 ºC vs. 15.4 ± 1.0 ºC; temperature range: 8.2 ± 1.2 ºC vs. 27.4 ± 2.9 ºC; P

≤ 0.03; Douglas-fir data in Table 1, lodgepole pine data reported in Griffin et al. (2011)).

Temperatures at 10-cm soil depth in undisturbed forests were not significantly different among

forest types, and averaged 10.9 ± 0.6 ºC. However, the range of 10-cm depth temperature was

lower in Douglas-fir (1.4 ± 0.6 ºC vs. 2.4 ± 0.2 ºC; P = 0.0236; Douglas-fir data in Table 1,

lodgepole pine data reported in Griffin et al. (2011)).

Undisturbed Douglas-fir and lodgepole pine forests also differed in several metrics of

litter, soil, and foliar N. Litter depth and all measurements of litter N content and pool size were

greater in Douglas-fir (Figure 3; P ≤ 0.03), though there was no difference in total litter mass. In

soils, extractable NH4+, NO3, total inorganic N, and total organic N were also greater in Douglas-

fir (P ≤ 0.005; Figure 4). Net N mineralization rates were not statistically different, however net

56

nitrification rates in undisturbed Douglas-fir were twice that of undisturbed lodgepole pine (4.8

vs 2.4 µg N g-1 day-1; Figure 4; P = 0.0242). Composite foliage in undisturbed forests did not

differ in foliar biomass, N content, or N pool size (Figure 5), however fresh foliar %N was

greater in Douglas-fir (P = 0.0185; Figure 5). P, K, Ca, and B in fresh foliage were also greater

in Douglas-fir, while Mn and Al were greater in lodgepole pine (P ≤ 0.03; Table 3).

Ecosystem response to bark beetle disturbance. Despite differing pre-disturbance

communities, the response of understory cover to bark beetle disturbance was similar between

forest types. Significant disturbance class effects were found for increased grass and forb cover

(P ≤ 0.04; Figure 2), but there were no significant interactions between species and disturbance

class for any biotic cover variable (Figure 2). However, soil temperature response did vary

among forest type. Significant disturbance class effects were found for decreases in the mean,

range, minimum, and maximum temperature at the litter-soil interface, though significant

interactions between species and disturbance class for the range and maximum show that

decreases in these variables were limited to lodgepole pine (P < 0.04; Douglas-fir data in Table

1; lodgepole pine data reported in Griffin et al. (2011)). No significant disturbance class or

species*disturbance class effects were found for 10-cm depth soil temperatures.

Outbreak-induced changes in N cycling also varied among forest types. Significant

effects of disturbance class were found for all metrics of litter quality and quantity except total

litter mass, which did not differ by either species or disturbance class (Figure 3). Nitrogen

concentration in both fresh and total litter also showed a significant interaction between species

and disturbance class, indicating that fresh litter %N increased only in lodgepole pine and total

litter %N increased only in Douglas-fir (Figure 3). In soils, significant disturbance class effects

were found for extractable NH4+ and NO3

-, net N mineralization, and net nitrification. No

57

significant effects were found for nitrification fraction, and there were no significant interactions

between species and disturbance class for any soil N metric (Figure 4). In trees unattacked by

bark beetles, all foliar N metrics showed a significant disturbance class effect except for fresh

foliar %N, which was marginally significant (Figure 5). N concentration of unattacked current

year foliage was higher in beetle-killed lodgepole pine, and composite foliar N was higher in

beetle-killed stands of both species. Live canopy biomass and canopy N pool size were lower in

beetle-killed stands of both types, but also showed a significant interaction between species and

disturbance class indicating that declines in these variables were greater in Douglas-fir compared

to lodgepole pine (Figure 5). Fresh foliar %N was positively related with net N mineralization in

soils of lodgepole pine (Adj. R2 = 0.56; P = 0.0202) but not in Douglas-fir.

Foliar micronutrients

Bark beetle disturbance significantly lowered foliar concentrations of Ca, Mg, and Mn in

fresh foliage (needles < 1yr) of unattacked Douglas-fir trees, while foliar concentrations of P, K,

Mg, S, Zn, B, Fe, Cu, Al and Na did not differ between Douglas-fir disturbance classes (Table

3). Foliar Mn also declined substantially in beetle-killed lodgepole pine forests, where decreases

in Al were also significant (Table 3). Forest types differed overall in foliar P, K, Ca, and S with

higher concentrations in Douglas-fir. Mn, Cu, and Al also differed by type, with greater

concentrations in lodgepole pine (Table 3). Across both forest types there was a significant

disturbance class effect for P, Mn and Al, with P increasing and Mn and Al decreasing in beetle-

killed stands relative to undisturbed (Table 3).

58

DISCUSSION

Bark beetle disturbance initiated similar changes in N cycling through litter and soil in both

Douglas-fir and lodgepole pine forests, despite substantial differences in pre-disturbance forest

structure and ecosystem N dynamics. Patterns of increased litter depth, needle litter N

concentration, and litter N pool size were consistent across forest types, though concurrent

increases in total litter %N were limited to Douglas-fir. In soils, patterns of increased extractable

inorganic N and N transformation rates following bark beetle disturbance were proportionally

similar among forest types, though Douglas-fir had more N-rich soils overall. Live foliar N pools

declined sharply with beetle-killed basal area in both forest types. Differences between these

foundation species (Ellison et al. 2005) in pre-disturbance ecosystem characteristics may explain

variable responses for some metrics following beetle outbreak. Total inorganic N in soils was

four-fold higher in Douglas-fir, and foliar N in undisturbed Douglas-fir averaged 1.23%

compared to 0.82% in lodgepole pine. These data suggest a higher overall N capital in Douglas-

fir forests and less potential N limitation. Thus, the response of foliar %N in unattacked trees to

increased soil N in beetle-killed stands was limited to lodgepole pine forests.

Forest types also differed in the response of abiotic soil conditions. Soil temperature was

cooler in beetle-killed lodgepole pine forests, presumably due to insulating effects of litter input

and increases in soil moisture. This effect was not seen in Douglas-fir, where the shading effect

of abundant understory cover may be the dominant control on soil temperature. Soil temperature

has been negatively correlated with net N mineralization across longer time scales of TSB in

lodgepole pine (Griffin et al. 2011), however neither forest type showed this relationship using

data from undisturbed and gray-stage TSB classes only. Furthermore, no litter N metrics were

correlated with net N mineralization in either forest type at this time-scale. Though unmeasured

59

in this study, soil moisture and soil C dynamics are also potential controls on soil N processes

and would be expected to change following bark beetle disturbance (Morehouse et al. 2008,

Spielvogel et al. 2009).

Observed changes in litter and soil N cycling of both forest types were consistent with

mechanisms of disturbance specific to the bark beetle insect guild. Rapid tree death appears to

inhibit the resorbtion of N from foliage prior to needlefall (Morehouse et al. 2008, Griffin et al.

2011), and deposits a large pulse of needle litter in a relatively short time. Though dead conifer

needles may leach some N compounds before dropping (Stadler et al. 2005), bark beetles do not

alter fluxes of inorganic N to soils in the same way as the frass of conifer defoliators (le Mellec

et al. 2009, Pitman et al. 2010). Beetle-induced inputs of N to the soil surface are largely in the

form of organic N in litter, which can be a N sink during a period of net N immobilization in the

early stages of decomposition (Fahey et al. 1985, Remsburg and Turner 2006). Thus,

disturbance-induced fluxes of N from canopy to soil pools may be spread over longer time

periods following bark beetle outbreak compared to defoliators. Minimal inputs of inorganic N to

the soil profile could explain why the magnitude of observed increases in net N mineralization

following bark beetle disturbance in lodgepole pine was smaller than those following defoliator

disturbance in another Pinus forest (Tokuchi et al. 2004).

Effects of bark beetle disturbance on micronutrient cycling could also be important for

post-beetle N cycling. Beetle-induced tree death reduces the delivery of photosynthate C to

ectomycorrhizal fungi. In lodgepole pine stands of the Yellowstone region, a similar loss of C

delivery to ectomycorrhizal communities following 50% defoliation has been shown to double

the level of manganese peroxidase activity in soils (Cullings et al. 2008), an extracellular fungal

enzyme used to metabolize lignin-based C sources. Greater demand for Mn to support increases

60

in this enzymatic pathway could reduce soil Mn availability to plants and may explain the

considerable declines in foliar Mn levels of unattacked, surviving trees in beetle-disturbed stands

of both forest types. In turn, litter produced from this foliage will also be lower in Mn content,

which is known to slow decomposition rates (Berg et al. 2007) and thus subsequent return of N

to soils from the litter layer. Furthermore, increased Mn peroxidase activity is likely to continue

for considerable time periods. Canopy foliage (an index of photosynthate C delivery to soil

fungi) in 30 yr post MBP lodgepole pine forest remains approximately half the biomass of

undisturbed forests (Griffin et al. 2011), and surviving trees may reduce C allocation to fungi

under conditions of elevated soil N (Hogberg et al. 2010). Concurrently, input of lignin and

woody C to soils increases with root turnover, more decomposing needle litter, and eventually

fine and coarse wood which has been shown to sequester large amounts of Mn during

decomposition (Daniel et al. 1997). Sustained declines in photosynthate C supply and increases

in lignin and woody C inputs could thus result in long-term declines of plant-available Mn.

Disturbance can lead to nutrient loss from forest ecosystems (Bormann and Likens 1979,

Chapin et al. 2004), however our data suggest three mechanisms of ecosystem N retention

following bark beetle outbreak. (1) Net N mineralization, net nitrification, and pool sizes of the

highly leachable NO3- ion remained low relative to other disturbance types in conifer ecosystems

(Turner et al. 2007, Tan et al. 2008), and nitrification fraction was unaffected. (2) Although the

foliar N pool declined in both forest types, increased foliar N of surviving trees partially

compensated for N lost in litterfall. (3) Declines in foliar Mn could affect future litter quality,

decreasing decomposition rate and slowing the return of inorganic N to the soil. Together, these

suggest that the potential for N loss as NO3- following bark beetle disturbance is low.

61

CONCLUSIONS

Forest types dominated by different foundation species (Ellison et al. 2005) can have

unique responses to a common disturbance type. N cycling was significantly impacted by bark

beetle outbreak in both Douglas-fir and lodgepole pine forests, which showed qualitatively

similar responses of increased litter N inputs and soil N processes. However, differences

between forest types in pre-disturbance ecosystem structure and N status resulted in variable

responses of soil temperature and foliar %N of unattacked trees. Though the magnitude of beetle

effects was relatively minor compared to other disturbance types, ecosystem impacts were

specific to disturbance by the bark beetle insect guild. Several mechanisms of ecosystem N

retention suggest that N losses following this native disturbance may be low.

ACKNOWLEDGEMENTS

We thank Gary Lovett, Bill Romme, and Dan Donato for constructive comments that

improved this manuscript. We also thank Emily Stanley, James Thoyre, and the Center for

Limnology at UW for access to laboratory equipment. For help in the field and the laboratory,

we appreciate the assistance of many University of Wisconsin students: Heather Lumpkin, Erin

Mellenthin, Alex Rahmlow, Amanda Rudie, and Ben Ruh.

REFERENCES

Berg, B., K. T. Steffen, and C. McClaugherty. 2007. Litter decomposition rate is dependent on litter Mn concentrations. Biogeochemistry 82:29-39. Binkley, D., R. Bell, and P. Sollins. 1992. Comparison of Methods for Estimating Soil-Nitrogen Transformations in Adjacent Conifer and Alder-Conifer Forests. Canadian Journal of Forest Research 22:858-863.

62

Boon, S. 2009. Snow ablation energy balance in a dead forest stand. Hydrological Processes 23:2600-2610. Bormann, F. H., and G. E. Likens. 1979. Pattern and process in a forested ecosystem: disturbance, development, and the steady state based on the Hubbard Brook ecosystem study. Springer-Verlag, New York. Bouyoucos, G. J. 1962. Hydrometer method improved for making particle and size analysis of soils. Agronomy Journal 54:464-465. Brown, J. K. 1978. Weight and density of crowns of Rocky Mountain conifers. USFS Intermountain Forest & Range Experimental Station Gen. Tech. Rep. INT-16. Byers, J. A. 1984. Electronic Multiprobe Thermometer and Multiplexer for Recording Temperatures of Microenvironments in the Forest Litter Habitat of Bark Beetles (Coleoptera, Scolytidae). Environmental Entomology 13:863-867. Chapin, F. S., H. A. Mooney, M. C. Chapin, and P. Matson. 2004. Principles of terrestrial ecosystem ecology. Springer. Cullings, K., G. Ishkhanova, and J. Henson. 2008. Defoliation effects on enzyme activities of the ectomycorrhizal fungus Suillus granulatus in a Pinus contorta (lodgepole pine) stand in Yellowstone National Park. Oecologia 158:77-83. Daniel, G., T. Nilsson, and J. Volc. 1997. Electron microscopical observations and chemical analyses supporting Mn uptake in white rot degraded Alstonia and pine wood stakes exposed in acid coniferous soil. Canadian Journal of Microbiology 43:663-671. Despain, D. G. 1990. Yellowstone Vegetation. Roberts Rinehart, Boulder. Ellison, A. M., M. S. Bank, B. D. Clinton, E. A. Colburn, K. Elliott, C. R. Ford, D. R. Foster, B. D. Kloeppel, J. D. Knoepp, G. M. Lovett, J. Mohan, D. A. Orwig, N. L. Rodenhouse, W. V. Sobczak, K. A. Stinson, J. K. Stone, C. M. Swan, J. Thompson, B. Von Holle, and J. R. Webster. 2005. Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3:479-486. Fahey, T. J., J. B. Yavitt, J. A. Pearson, and D. H. Knight. 1985. The Nitrogen-Cycle in Lodgepole Pine Forests, Southeastern Wyoming. Biogeochemistry 1:257-275. Giardina, C. P., M. G. Ryan, R. M. Hubbard, and D. Binkley. 2001. Tree species and soil textural controls on carbon and nitrogen mineralization rates. Soil Science Society of America Journal 65:1272-1279. Griffin, J. M., M. G. Turner, and M. Simard. 2011. Nitrogen cycling following mountain pine beetle disturbance in lodgepole pine forests of Greater Yellowstone. Forest Ecology and Management 261:1077-1089.

63

Haack, R. A., and J. W. Byler. 1993. Insects and Pathogens - Regulators of Forest Ecosystems. Journal of Forestry 91:32-37. Hais, M., and T. Kucera. 2008. Surface temperature change of spruce forest as a result of bark beetle attack: remote sensing and GIS approach. European Journal of Forest Research 127:327-336. Hawkins, B. J., H. Boukcim, and C. Plassard. 2008. A comparison of ammonium, nitrate and proton net fluxes along seedling roots of Douglas-fir and lodgepole pine grown and measured with different inorganic nitrogen sources. Plant Cell and Environment 31:278- 287. Hogberg, M. N., M. J. I. Briones, S. G. Keel, D. B. Metcalfe, C. Campbell, A. J. Midwood, B. Thornton, V. Hurry, S. Linder, T. Nasholm, and P. Hogberg. 2010. Quantification of effects of season and nitrogen supply on tree below-ground carbon transfer to ectomycorrhizal fungi and other soil organisms in a boreal pine forest. New Phytologist 187:485-493. Keane, R. E. 2008. Biophysical controls on surface fuel litterfall and decomposition in the northern Rocky Mountains, USA. Canadian Journal of Forest Research 38:1431-1445. Koukol, O., B. Benova, M. Vosmanska, T. Frantik, M. Vosatka, and M. Kovarova. 2008. Decomposition of spruce litter needles of different quality by Setulipes androsaceus and Thysanophora penicillioides. Plant and Soil 311:151-159. le Mellec, A., M. Habermann, and B. Michalzik. 2009. Canopy herbivory altering C to N ratios and soil input patterns of different organic matter fractions in a Scots pine forest. Plant and Soil 325:255-262. Lovett, G. M., C. D. Canham, M. A. Arthur, K. C. Weathers, and R. D. Fitzhugh. 2006. Forest ecosystem responses to exotic pests and pathogens in eastern North America. Bioscience 56:395-405. Lovett, G. M., K. C. Weathers, M. A. Arthur, and J. C. Schultz. 2004. Nitrogen cycling in a northern hardwood forest: Do species matter? Biogeochemistry 67:289-308. McMillin, J. D., and K. K. Allen. 2003. Effects of Douglas-fir beetle (Coleoptera : Scolytidae) infestations on forest overstory and understory conditions in western Wyoming. Western North American Naturalist 63:498-506. Morehouse, K., T. Johns, J. Kaye, and A. Kaye. 2008. Carbon and nitrogen cycling immediately following bark beetle outbreaks in southwestern ponderosa pine forests. Forest Ecology and Management 255:2698-2708.

64

Negron, J. F. 1998. Probability of infestation and extent of mortality associated with the Douglas-fir beetle in the Colorado Front Range. Forest Ecology and Management 107:71-85. Pitman, R. M., E. I. Vanguelova, and S. E. Benham. 2010. The effects of phytophagous insects on water and soil nutrient concentrations and fluxes through forest stands of the Level II monitoring network in the UK. Science of the Total Environment 409:169-181. Prescott, C. E. 2005. Do rates of litter decomposition tell us anything we really need to know? Forest Ecology and Management 220:66-74. Prescott, C. E., H. N. Chappell, and L. Vesterdal. 2000. Nitrogen turnover in forest floors of coastal Douglas-fir at sites differing in soil nitrogen capital. Ecology 81:1878-1886. Przybyl, K., P. Karolewski, J. Oleksyn, A. Labedzki, and P. B. Reich. 2008. Fungal diversity of Norway spruce litter: Effects of site conditions and premature leaf fall caused by bark beetle outbreak. Microbial Ecology 56:332-340. Raffa, K. F., and A. A. Berryman. 1987. Interacting Selective Pressures in Conifer-Bark Beetle Systems - a Basis for Reciprocal Adaptations. American Naturalist 129:234-262. Remsburg, A. J., and M. G. Turner. 2006. Amount, position, and age of coarse wood influence litter decomposition in postfire Pinus contorta stands. Canadian Journal of Forest Research 36:2112-2123. Safranyik, L., and B. Wilson, editors. 2006. The Mountain Pine Beetle: A Synthesis of Biology, Management, and Impacts on Lodgepole Pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, BC. SAS Institute Inc. 2003. SAS/STAT Statistical Software, v. 9.1. 3rd edition. SAS Institute, Inc., Cary, NC. Schowalter, T. D. 1981. Insect Herbivore Relationship to the State of the Host Plant - Biotic Regulation of Ecosystem Nutrient Cycling through Ecological Succession. Oikos 37:126- 130. Scott, N. A., and D. Binkley. 1997. Foliage litter quality and annual net N mineralization: Comparison across North American forest sites. Oecologia 111:151-159. Shore, T. L., L. Safranyik, B. C. Hawkes, and S. W. Taylor. 2006. Effects of the mountain pine beetle on lodgepole pine stand structure and dynamics. Pages 95-114 in L. Safranyik and B. Wilson, editors. The mountain pine beetle: A synthesis of biology, management, and impacts on lodgepole pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, B.C.

65

Spielvogel, S., J. Prietzel, K. Auerswald, and I. Kogel-Knabner. 2009. Site-specific spatial patterns of soil organic carbon stocks in different landscape units of a high-elevation forest including a site with forest dieback. Geoderma 152:218-230. Stadler, B., T. Muller, D. Orwig, and R. Cobb. 2005. Hemlock woolly adelgid in new england forests: Canopy impacts transforming ecosystem processes and landscapes. Ecosystems 8:233-247. Tan, X., S. X. Chang, P. G. Comeau, and Y. H. Wang. 2008. Thinning effects on microbial biomass, N mineralization, and tree growth in a mid-rotation fire-origin lodgepole pine stand in the lower foothills of Alberta, Canada. Forest Science 54:465-474. Thomas, K. D., and C. E. Prescott. 2000. Nitrogen availability in forest floors of three tree species on the same site: the role of litter quality. Canadian Journal of Forest Research:1698-1706. Tokuchi, N., N. Ohte, S. Hobara, S. J. Kim, and K. Masanori. 2004. Changes in biogeochemical cycling following forest defoliation by pine wilt disease in Kiryu experimental catchment in Japan. Hydrological Processes 18:2727-2736. Turner, M. G., E. A. H. Smithwick, K. L. Metzger, D. B. Tinker, and W. H. Romme. 2007. Inorganic nitrogen availability following severe stand-replacing fire in the Greater Yellowstone Ecosystem. Proceedings of the National Acadamy of Science 104:4782- 4789. USFS. 2006. Forest Insect and Disease Survey, regions 1, 2, & 4. UWSPAL. 2010. Soil and Plant Analysis.in J. Peters, editor. Wisconsin Procedures for Soil Testing, Plant Analysis and Feed & Forage Analysis. University of Wisconsin Soil and Plant Analysis Lab, Madison, WI.http://uwlab.soils.wisc.edu/procedures.html. WRCC. 2010. Moran, WY, climate summary. Western Regional Climate Center. http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?wy6440.

66

TABLES

Table 1. Site characteristics in undisturbed and bark beetle-killed Douglas-fir forests. N = 5 /

disturbance class. Significant (α < 0.05 ) ANOVA P values are shown in bold.

Douglas-fir disturbance class Site characteristic Undisturbed Beetle-killed ANOVA P Topography Elevation (m) 2186 ± 38 2325 ± 107 0.0400 Aspect (SW index) 0.367 ± 0.691 -0.085 ± 0.566 0.9201 Slope 26 ± 12 22 ± 11 0.6282 Canopy vegetation Basal area (m2 ha-1) Total 62.7 ± 9.9 75.3 ± 12.2 0.1468 Douglas-fir 58.7 ± 12.6 75.1 ± 12.1 0.0973 Live Douglas-fir 57.6 ± 13.2 5.2 ± 2.9 0.0001 Dead Douglas-fir 1.1 ± 1.2 69.9 ± 11.2 <0.0001 Live foliar biomass (kg ha-1) 15,179 ± 4021 2551 ± 1538 0.0004 Soils Growing season temperature Litter-soil interface (ºC) mean 13.5 ± 0.7 12.7 ± 2.0 0.5073 range 8.2 ± 1.2 9.4 ± 2.8 0.4607 maximum 18.1 ± 1.0 18.3 ± 3.7 0.9094 minimum 9.9 ± 0.8 8.9 ± 1.0 0.1953 10 cm soil depth (ºC) mean 11.2 ± 1.1 10.5 ± 1.6 0.4899 range 1.4 ± 0.6 1.5 ± 0.3 0.7494 maximum 11.9 ± 1.3 11.3 ± 1.7 0.5892 minimum 10.5 ± 0.8 9.8 ± 1.5 0.4284 Texture & Structure Sand (%) 56 ± 3 49 ± 3 0.1590 Silt (%) 33 ± 6 36 ± 5 0.3570 Clay (%) 12 ± 2 15 ± 2 0.0847 Bulk density (g cm-3) 0.66 ± 0.22 0.63 ± 0.10 0.8195 Organic soil depth (cm) 4.2 ± 1.5 5.5 ± 2.0 0.3078 Organic matter (%) 8.4 ± 4.8 8.6 ± 2.2 0.9299 Cations pH 6.2 ± 0.3 6.4 ± 0.2 0.2073 Ca (µg g-1; exch.) 2401 ± 808 2702 ± 253 0.4974 Mg (µg g-1; exch.) 187 ± 52 220 ± 14 0.2527 K (µg g-1; exch.) 225 ± 79 218 ± 42 0.8696 P (µg g-1; exch.) 49 ± 8.4 51 ± 15 0.7950 Litter Depth (cm) 3.0 ± 0.5 3.5 ± 0.6 0.2968 Mass (g m-2) 1760 ± 365 1924 ± 357 0.5385

67

Table 2. N concentrations, pools, and fluxes in undisturbed and bark beetle-killed Douglas-fir

forests. N = 5 / disturbance class; values in table are untransformed mean ± 2SE. Significant (α <

0.05 ) ANOVA P values are shown in bold; asterisk (*) denotes marginally significant (0.05 < α

< 0.07) ANOVA P.

1 µg N g soil-1 2 µg N g soil-1 yr-1 3 ANOVA performed on log(x) transformed data

4 ANOVA performed on 1/x transformed data

Douglas-fir disturbance class N concentration, pool, or flux Undisturbed Beetle-killed ANOVA P Litter Total litter N pool (g N m-2) 19.3 ± 3.3 26.4 ± 5.8 0.0666* Total litter %N 1.10 ± 0.06 1.37 ± 0.13 0.0054 Total litter C:N 44 ± 2 35 ± 3 0.0008 All needles %N 0.89 ± 0.07 1.38 ± 0.10 <0.0001 All needles C:N 57 ± 5 36 ± 2 <0.0001 < 1 yr old needles %N 0.70 ± 0.06 0.79 ± 0.07 0.0641* < 1 yr old needles C:N 72 ± 6 62 ± 5 0.0266

Soil Total soil N pool (g N m-2) 192 ± 78 237 ± 42 0.3353 %N 0.23 ± 0.15 0.26 ± 0.07 0.7782 C:N 19.2 ± 2.9 16.7 ± 0.4 0.11094 Organic N (mg N g-1) 2.34 ± 1.52 2.58 ± 0.70 0.7836 NH4

+ 1 2.5 ± 0.6 5.0 ± 0.8 0.0010 NO3

- 1 1.5 ± 1.0 5.5 ± 5.2 0.0520*3 Total inorganic N 1 4.0 ± 1.7 10.5 ± 4.7 0.00603 Net N mineralization 2 14.3 ± 9.3 24.8 ± 11.2 0.09503 Net nitrification 2 4.8 ± 0.6 7.2 ± 1.0 0.0295 Nitrification fraction 0.42 ± 0.18 0.34 ± 0.11 0.4724 Canopy foliage Foliar N pool (g N m-2) 11.5 ± 1.7 2.4 ± 1.5 0.0004 All needle %N 0.76 ± 0.04 0.90 ± 0.07 0.0101 All needle C:N 63 ± 3 54 ± 4 0.0046 <1yr needle %N 1.23 ± 0.19 1.34 ± 0.12 0.3853 <1yr needle C:N 40 ± 7 36 ± 3 0.3499

68

Table 3. Foliar chemistry of new needles from unattacked trees in undisturbed and beetle-killed stands of Douglas-fir and lodgepole

pine in the Greater Yellowstone Ecosystem. N = 5 stands per disturbance class; values in table are untransformed mean ± 2SE.

Significant (α < 0.05 ) ANOVA P values are shown in bold; asterisk (*) denotes marginally significant (0.05 < α < 0.06) ANOVA P.

1 ANOVA performed on log10(x) transformed data 2 ANOVA performed on x-2 transformed data 3 ANOVA performed on √x transformed data 4 ANOVA performed on 1/x transformed data 5 ANOVA performed on 1/x2 transformed data

Douglas-fir Lodgepole pine Species Comparison ANOVA P Foliar Nutrient Undisturbed Beetle-killed ANOVA P Undisturbed Beetle-killed ANOVA P Species Class Species*Class N (%) 1.23 ± 0.19 1.34 ± 0.12 0.3853 0.82 ± 0.03 1.00 ± 0.06 0.0048 <0.0001 0.0578* 0.6043 C:N 39.7 ± 7.1 35.8 ± 3.3 0.3499 59.1 ± 3.9 49.5 ± 2.4 0.0044 < 0.0001 0.0154 0.2613 N:P 5.5 ± 0.4 5.3 ± 0.4 0.5771 6.1 ± 0.2 6.4 ± 0.5 0.4200 0.0010 0.7615 0.3123 P (%) 0.23 ± 0.03 0.25 ± 0.03 0.2099 0.14 ± 0.00 0.16 ± 0.01 0.0555* < 0.0001 0.0426 0.9067 1 K (%) 1.2 ± 0.1 1.1 ± 0.1 0.4180 0.50 ± 0.04 0.58 ± 0.04 0.5153 < 0.0001 0.3032 0.9785 1 Ca (%) 0.40 ± 0.05 0.31 ± 0.04 0.0236 0.14 ± 0.04 0.15 ± 0.04 0.8026 < 0.0001 0.1104 0.0565* Mg (%) 0.11 ± 0.00 0.10 ± 0.01 0.0344 0.09 ± 0.01 0.10 ± 0.01 0.4742 0.2624 0.9954 0.1255 S (%) 0.09 ± 0.01 0.10 ± 0.01 0.7073 0.07 ± 0.00 0.08 ± 0.01 0.8023 0.0011 0.6660 0.8831 Zn (ppm) 28.5 ± 4.3 27.5 ± 4.0 0.7352 31.9 ± 1.5 31.5 ± 9.4 0.4993 5 0.2523 0.6608 0.9437 1 B (ppm) 13.6 ± 0.7 13.7 ± 2.8 0.9178 12.0 ± 0.9 11.9 ± 3.1 0.9644 0.1836 0.9822 0.9203 Mn (ppm) 132 ± 45 67 ± 22 0.0331 340 ± 18 200 ± 34 0.0010 < 0.0001 < 0.0001 0.0537* Fe (ppm) 44.4 ± 28.4 33.7 ± 7.9 0.5810 4 32.0 ± 2.4 29.1 ± 3.5 0.2780 0.3743 0.4461 0.6734 2 Cu (ppm) 7.3 ± 3.1 7.7 ± 2.5 0.8277 2.3 ± 0.6 3.1 ± 0.8 0.1714 0.0001 0.2954 0.5322 1 Al (ppm) 52.0 ± 41.9 18.8 ± 8.4 0.0780 1 492 ± 114 329 ± 48 0.0217 < 0.0001 0.0048 0.4533 3 Na (ppm) 12.4 ± 4.5 11.2 ± 1.4 0.6255 8.7 ± 3.2 11.3 ± 3.4 0.3455 0.3085 0.6951 0.2966

68

69

FIGURE LEGENDS

Figure 1. Hypothesized impacts of bark beetle disturbance in Douglas-fir and lodgepole

pine ecosystems. Bark beetle-induced tree mortality is predicted to increase litter N inputs and

moderate soil temperature. Both of these changes may accelerate fluxes of inorganic soil N and

increase pools of plant-available N. Unattacked trees within beetle-killed stands may respond to

these increases by allocating more N to foliage. Forest types are expected to differ in pre-

disturbance N pool sizes and flux rates, which may result in variable ecosystem responses.

Figure 2. Tree basal area (A) and understory cover by plant functional group (B) in

undisturbed and bark beetle-killed lodgepole pine and Douglas-fir forests. Plotted values are

untransformed mean ± 2SE. LP and DF denote a significant effect of disturbance class within

lodgepole pine (LP) or Douglas-fir (DF) forests only (P < 0.05; N = 10; GLM: y = disturbance

class). SPP, CL, and S*C denote significant (P < 0.05; N = 20) effects of species (SPP),

disturbance class (CL), and their interaction (S*C) in the global GLM: y = species + disturbance

class + species*disturbance class. Total basal area: SPP, S*C. Total cover: LP, DF, SPP, CL.

Figure 3. Litter quantity and litter N metrics in undisturbed and bark beetle-killed

lodgepole pine and Douglas-fir forests. Plotted values are untransformed mean ± 2SE. (A)

litter depth, (B) litter mass, (C) fresh (<1 yr) needle litter %N, (D) all needle litter %N, (E) total

litter %N, (F) total litter N pool. Values inset within each panel are P values from the GLM: y =

species + disturbance class + species*disturbance class.

70

Figure 4. Soil N metrics in undisturbed and bark beetle-killed lodgepole pine and Douglas-

fir forests. Plotted values are untransformed mean ± 2SE. (A) extractable NH4+, (B) extractable

NO3-, (C) net N mineralization, (D) net nitrification, (E) nitrification fraction, (F) organic N.

Values inset within each panel are P values from the GLM: y = species + disturbance class +

species*disturbance class.

Figure 5. Canopy foliar N metrics and foliar biomass in undisturbed and bark beetle-killed

lodgepole pine and Douglas-fir forests. Plotted values are untransformed mean ± 2SE. (A)

fresh foliar %N, (B) total foliar %N, (C) live canopy biomass, (D) canopy N pool. Values inset

within each panel are P values from the GLM: y = species + disturbance class +

species*disturbance class.

71

FIGURES

Figure 1

Similar effects; increased %N in canopy foliage and decreased N pool size in beetle-killed stands of both species.

Increased %N in canopy foliage.Sharp decline in N pool size.

Canopy of unattacked trees

Larger increases in pool size and flux rate in Douglas-fir; greater soil N in Douglas-fir overall

Increased NH4+, NO3

-, net N mineralization, & net nitrification

Soils

Greater increases in litter %N and N pool size in Douglas-fir; greater litter N in Douglas-fir overall

Increased litter mass, depth, and %N of fresh needle litter

Litter

Increased understory cover in beetle-killed forests of both types, with more cover overall in Douglas-fir.

Increased cover-canopy opening-increased light, H2O, nutrients

Understory cover

Similar effects; cooler soil temperatures in beetle-killed forests of both species

Decreased soil temperature-insulating litter input -shading by understory growth

Soil temperature

Hypothesized differences in beetle effects in Douglas-fir compared to lodgepole pine forests

Hypothesized bark beetle effects in Douglas-fir forests

Ecosystem component

Similar effects; increased %N in canopy foliage and decreased N pool size in beetle-killed stands of both species.

Increased %N in canopy foliage.Sharp decline in N pool size.

Canopy of unattacked trees

Larger increases in pool size and flux rate in Douglas-fir; greater soil N in Douglas-fir overall

Increased NH4+, NO3

-, net N mineralization, & net nitrification

Soils

Greater increases in litter %N and N pool size in Douglas-fir; greater litter N in Douglas-fir overall

Increased litter mass, depth, and %N of fresh needle litter

Litter

Increased understory cover in beetle-killed forests of both types, with more cover overall in Douglas-fir.

Increased cover-canopy opening-increased light, H2O, nutrients

Understory cover

Similar effects; cooler soil temperatures in beetle-killed forests of both species

Decreased soil temperature-insulating litter input -shading by understory growth

Soil temperature

Hypothesized differences in beetle effects in Douglas-fir compared to lodgepole pine forests

Hypothesized bark beetle effects in Douglas-fir forests

Ecosystem component

72

Figure 2

0

20

40

60

80

100

120

Undisturbed Beetle-killed Undisturbed Beetle-killed

% C

over

Tree seedlings: n.s.Forbs: LP, DF, SPP, CLShrubs: SPPSedges: n.s.Grasses: DF, SPP, CL

Lodgepole pine Douglas-fir

0102030405060708090

Bas

al a

rea

(m2 h

a-1)

Dead: LP, DF, SPP, CL, S*C

Live: LP, DF, SPP, S*CA)

B)

73

Figure 3

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5Li

tter d

epth

(cm

) A) Species: 0.002

Class: 0.057Species*Class: ns

0

500

1000

1500

2000

2500

3000

Litte

r mas

s (g

m-2

)

Species: nsClass: nsSpecies*Class: ns

B)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Fres

h ne

edle

litte

r %N

Species: 0.002Class: <0.001Species*Class: 0.045

C)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

All

need

le li

tter %

N

Species: <0.001Class: <0.001Species*Class: ns

D)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

PICO PSME

Tota

l litt

er %

N

Species: <0.001Class: 0.003Species*Class: 0.021

E)

0

5

10

15

20

25

30

35

PICO PSME

Tota

l litt

er N

poo

l siz

e (g

N m-2

)

Species: 0.002Class: 0.035Species*Class: ns

F)

74

Figure 4

0

2

4

6

8

10

12E

xtra

ctab

le N

H4+ (µ

g N

g s

oil-1

)A)

Species: <0.001Class: 0.001Species*Class: ns

0

2

4

6

8

10

12

Ext

ract

able

NO3

- (µg

N g

soi

l-1)

Species: <0.001Class: 0.053Species*Class: ns

B)

0

5

10

15

20

25

30

35

40

Net

N m

iner

aliz

atio

n (µ

g N

g s

oil-1 y

r-1)

Species: nsClass: 0.012Species*Class: ns

C)

0123456789

10

Net

nitr

ifica

tion

(µg

N g

soi

l-1 y

r-1)

Species: 0.027Class: 0.023Species*Class: ns

D)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

PICO PSME

Nitr

ifica

tion

fract

ion

Species: nsClass: nsSpecies*Class: ns

E)

0

500

1000

1500

2000

2500

3000

3500

4000

4500

PICO PSME

Org

anic

N (µ

g N

g s

oil-1

)

Species: <0.001Class: nsSpecies*Class: ns

F)

75

Figure 5

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

PICO PSME

Fres

h fo

liar %

NSpecies: >0.001Class: 0.058Species*Class: ns

A)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

PICO PSME

Tota

l fol

iar %

N

Species: nsClass: 0.0003Species*Class: ns

B)

0

20

40

60

80

100

120

140

160

PICO PSME

Can

opy

N p

ool (

kg N

ha-1

)

Species: ns Class: <0.0001Species*Class: 0.040

D)

02000400060008000

100001200014000160001800020000

PICO PSME

Live

can

opy

biom

ass

(kg

ha-1

)

Species: ns Class: <0.001Species*Class: 0.045

C)

76

CHAPTER 3

Does salvage logging alter ecosystem response to

bark beetle disturbance in lodgepole pine forests?

ABSTRACT

Unprecedented outbreaks of native bark beetles in conifer forests of western North

America have elicited multiple management strategies including salvage logging. Salvage has

the potential to impact a wider range of forest ecosystem components than beetle outbreak alone,

which may lead to different ecosystem responses in salvaged forests compared to beetle-killed

only. In this study we use a paired and replicated before-after-control-impact (BACI) design to

test the first-year effects of salvage on sapling density, understory cover, soil microclimate, and

litter and soil N cycling of lodgepole pine (Pinus contorta Dougl.) forests attacked by the

mountain pine beetle (Dendroctonus ponderosae Hopkins). Salvage harvest removed most basal

area and decreased the density of all tree saplings (height <1.4 m) by 50%, although lodgepole

pine sapling density was unchanged and appeared adequate for stand replacement. Understory

biotic cover declined with salvage harvest, graminoid and shrub cover was unchanged, and cover

of downed coarse wood increased. Mid-summer soil temperature increased by ~1 ºC at both the

litter-soil interface and 10 cm depth in salvaged plots, and the N concentration of needle litter

increased by 30%. Soil and resin bag N were largely unaffected by salvage except for extractable

soil NO3- which tripled in concentration but remained low at 1.5 µg N g soil-1. Collectively, the

initial effects of post-beetle salvage were modest. Beetle-induced declines in plant N uptake and

increases in litter input appeared to be the main drivers of change in litter and soil N cycling over

time, with little additional influence of salvage during the first year following harvest.

77

INTRODUCTION

Concurrent outbreaks of several native bark beetle species throughout western North

America in the past decade have led to concern regarding ecosystem function, timber production,

tourism, and forest fire (Flint 2006, Flint and Haynes 2006, McFarlane and Witson 2008). In

response, management strategies have utilized both preemptive and post-outbreak approaches to

mediate the impact of this disturbance. Preemptive strategies include disrupting or mimicking

beetle pheromone communication (Bentz et al. 2005, Progar 2005, Gillette et al. 2006, Jeans-

Williams and Borden 2006), and using silvicultural methods to reduce stand susceptibility

(Safranyik et al. 1999, Fettig et al. 2007). Both methods have had moderate successes over

limited temporal and spatial scales, yet landscape-scale suppression of beetle outbreak is unlikely

(Raffa et al. 2008). Post-outbreak remediation techniques include fuel reduction treatments and

salvage harvests. Because beetle and salvage logging disturbances affect different ecosystem

components, interactions of these disturbance types may result in unique consequences for

ecosystem structure and function (Paine et al. 1998) compared to bark beetles alone.

Canopy biomass is a strong control on the rates and retention efficiency of forest N

cycling (Prescott 2002). Standing dead wood contains a significant amount of N (Fahey et al.

1985, Edmonds and Eglitis 1989, Herrmann and Prescott 2008) that is exported from the

ecosystem by salvage. However, rates of N input from coarse wood are small and contribute

little to annual rates of soil N turnover (Laiho and Prescott 2004). Fine woody debris and needle

litter (needles, twigs, reproductive structures, and bark) have higher N content and faster

turnover than coarse wood, and thus contribute more to overall N cycling in conifer forests

(Laiho and Prescott 1999). Beetle-induced tree mortality initiates a pulse of N-enriched needle

litter as the canopy decays (Morehouse et al. 2008, Klutsch et al. 2009, Griffin et al. 2011).

78

Salvage inputs of litter and fine woody debris are variable depending on method and site

treatment (Smethurst and Nambiar 1990, Goodman and Hungate 2006), and in beetle-killed

stands would likely depend on the extent of canopy deterioration that occurred due to beetles

prior to salvage. Litter and slash material may serve as a temporary N sink if it remains on the

forest floor undisturbed rather than burned, however slash burning can cause a pulse of soil

inorganic N lasting up to three years (Lopushinsky et al. 1992). Slash effects on soil N also vary

with time since disturbance and forest type. Following an outbreak of spruce beetle

(Dendroctonus rufipennis Kirby) in Alaska, Goodman and Hungate (2006) found no effect of

salvage on soil inorganic N four years after harvest. In contrast, a variety of slash treatments

following non-salvage harvest of Monterey pine (Pinus radiata D. Don.) forests all yielded

elevated N mineralization in the first two years after harvest (Smethurst and Nambiar 1990).

Logging activity can have a strong influence on both abiotic and biotic properties of

forest soils. Machinery can cause soil compaction and increased bulk density (Blouin et al. 2005)

that reduces porosity and water holding capacity, and increases soil strength (Blouin et al. 2008).

Soil temperature may also be elevated by logging, though the magnitude is dependent upon site-

treatment methods (Smethurst and Nambiar 1990). In contrast, beetle-induced litterfall can

decrease soil temperatures in the first few years after attack (Griffin et al. 2011). In turn, these

abiotic changes contribute to shifts in microbial community structure (Chatterjee et al. 2008) and

mycorrhizal associations (Kranabetter et al. 2006). The net result of these changes on soil N

dynamics is unclear. Some studies report short-lived increases in mineralization rates following

compaction (Kranabetter et al. 2006); others report reductions in soil inorganic N pool sizes

(Blouin et al. 2005, Choi et al. 2005, Kamaluddin et al. 2005). Variability among results may be

79

due to differing degrees of compaction, variation in slash treatments, or variation among forest

types in nutrient content of ecosystem pools.

Beetle and salvage disturbances also differ in their potential impact on tree saplings (here

defined as stems <1.4 m high), which provide a source of advance regeneration that could

ultimately lead to canopy replacement. In lodgepole pine (Pinus contorta Dougl.) forests,

successful mountain pine beetle (Dendroctonus ponderosae Hopkins) attack is limited to canopy

and large sub-canopy trees (Safranyik and Carroll 2006). Thus, post-beetle sapling densities are

largely unchanged and are usually sufficient for stand replacement with species compositions

consistent with local substrate and site conditions (Shore et al. 2006, Astrup et al. 2008, Nigh et

al. 2008, Rocca and Romme 2009, Vyse et al. 2009). However, salvage logging can have direct

and indirect, positive and negative effects on sapling density (Jonasova and Prach 2004). Direct

effects include physical damage or mortality of existing seedlings and saplings (Donato et al.

2006). Indirect mechanisms involve the effects of soil compaction and nutrient levels on tree

seedling growth and survival (Blouin et al. 2005, Bulmer and Simpson 2005, Kamaluddin et al.

2005, Blouin et al. 2008) and alteration of microsite conditions such as bare mineral soil or nurse

logs necessary for successful seedling establishment (Jonasova and Prach 2004). A flush of

understory growth and community composition change are common responses to both beetle

(McCambridge et al. 1982, Stone and Wolfe 1996, McMillin and Allen 2000) and salvage

(Hanson and Stuart 2005, del Rio 2006) disturbances, and are likely driven by increased light,

water, nutrient, and substrate availability.

In turn, litter inputs and vegetative cover can affect soil N cycling by shading and cooling

soils, providing N sinks in biomass, or altering rates of N fixation. Studies of other forest

management techniques suggest that soil disturbance is a key driver of vegetation differences

80

among beetle-killed stands with and without salvage. In cut-and-leave treatments, beetle-killed

trees are felled by hand and left on-site, requiring no heavy machinery or log skidding that

disturbs soil. Following an outbreak of the southern pine beetle (Dendroctonus frontalis

Zimmerman) in loblolly pine (Pinus taeda (L.)), Coleman et al. (2008) showed few differences

in understory tree or ground vegetation responses between beetle-killed stands with and without

this treatment type.

In this study, we use field sampling and laboratory analyses to compare stand structure,

ground cover, soil temperature, litter quantity and quality, and soil N metrics in beetle-killed vs.

beetle-killed + salvage logged stands of lodgepole pine. We hypothesized that salvage would

reduce sapling density, and increase the cover of bare soil, coarse wood, and understory

vegetation. We also hypothesized litter mass, depth, and N pool size would increase as more

material was added to the forest floor following salvage. Expectations for soil temperature were

less clear, as canopy opening would tend to increase soil temperatures while additional inputs of

litter mass may serve to insulate soils and cool soil temperature. With the addition of soil

disturbance from logging in beetle-killed forests, we expected soil inorganic N pools and the

accumulation of N on buried resin bags to also increase relative to beetle disturbance alone.

METHODS

Study region and experimental design

All study sites were located within a 4 km2 area of the Green River Lakes region on the

Bridger-Teton National Forest in northeastern Wyoming, USA (Figure 1). Forests are dominated

by lodgepole pine, with minor components of subalpine fir (Abies lasiocarpa Hook.), Engelmann

spruce (Picea engelmannii Parry), and whitebark pine (Pinus albicaulis Engelmann) (Despain

81

1990). Mean temperature is 14.2 °C July and -10.2 °C in January, and the region averages 297

mm precipitation per year, mostly as snow (WRCC 2010a). Soils are relatively nutrient poor and

derived from andesitic substrates. Mountain pine beetle activity in the area peaked in 2005,

though affected forests were of a mix of unattacked trees and beetle-killed trees in the red and

gray stages of canopy decline (Griffin et al. 2011) at the time of initial sampling in 2007.

We used a paired and replicated before-after-control-impact (BACI) experimental design

(Underwood 1994) to test for an effect of salvage-logging on stand structure, ground cover, soil

microclimate, and N cycling parameters in litter and soils of beetle-killed lodgepole pine. Pre-

salvage structure, litter, and soil N pool data were collected in 2007 from 16 circular plots of 8 m

radius. Eight plots were located in beetle-killed stands designated to be salvage-logged, and

eight were located in paired similar beetle-killed stands <400m away not designated for salvage.

Salvage logging in the beetle + salvage plots occurred in summer 2009. Harvest included both

beetle-killed and unattacked trees, though >90% of the basal removed was beetle-killed. Cutting

was performed by feller-buncher, with de-limbing done at landing sites and slash returned to and

scattered in the plots. Post-salvage structure, litter, and soil N pool data were collected from all

16 plots in summer 2010, while soil temperatures and resin bag N accumulation were measured

continuously from summer 2008 through summer 2010.

Vegetation

All live and dead trees greater than breast height (1.4m) were identified to species and

measured for diameter at breast height (DBH). Beetle-killed trees were identified by the presence

of exit holes and pitch tubes on the bark exterior, boring dust at the base of trees, and distinctive

J-shaped Dendroctonus galleries underneath the bark. Trees <1.4m were tallied by species and

measured to the nearest 10-cm height class in the northeast quadrant of each 8-m radius plot

82

(50m2), as well as in two 50-m2 rectangular plots 15 m to the east and west. Ground cover of

litter, coarse wood, bare soil, and three plant functional groups (forbs, sedges, and graminoids

(grasses + sedges)) was visually estimated to the nearest 10% in ten 0.25-m2 circular microplots;

total cover was allowed to exceed 100% to account for multiple strata of ground vegetation.

Microplots were located within the central 5-m radius area of the plot using a stratified random

design of fixed distances (one at 0.5 m; two at 1.5 m, 2.5 m and 3.5 m; and three at 4.5 m) and

random bearings (in 10º increments) from the plot center.

Microclimate

Soil temperature was measured hourly in four of the eight plot pairs from June 2008

through September 2010 using three pairs of iButton datalogger probes (Maxim Integrated

Products Inc., Dallas Semiconductor, Sunnyvale, CA) per plot. Beetle + salvage plots were not

instrumented during the salvage period in summer 2009. One of each iButton pair was installed

at the litter-organic soil interface, and the second was installed at 10 cm soil depth. For each

depth, temperature probe data were summarized as follows. First, plot-level hourly temperatures

were determined by averaging hourly data from the three probes per plot. Plot-level daily mean

temperatures were then calculated from the hourly data, and averaged to determine treatment-

level daily means.







83


was composited by plot for each category, and ground to powder for C:N analysis on a Leco

CNS-2000 at the University of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL 2010).

Soil chemistry, soil N pools, and resin bag N

One soil core was collected from each 0.25-m2 microplot (N = 10 per plot) using a 5-cm

diameter x 15-cm long PVC corer. Soils were sieved (2 mm mesh), weighed, and divided into

three subsamples: (1) 30 g oven-dried at 60 ºC for gravimetric percent moisture (2007 and 2010);

(2) 20 g extracted in 75 ml of 2M KCl for 2 hours, with the extract then filtered and frozen for

later analysis of NH4+ and NO3

- pools (2007 and 2010); (3) 20 g air-dried and bulked by plot for

soil texture and chemical analyses (2007 only). Air-dried soil was analyzed for pH, total N,

exchangeable Ca, Mg, and K, available P (Bray P1 extract), and organic matter at the University

of Wisconsin Soil and Plant Analysis Laboratory (UWSPAL 2010). Soil organic N was

determined by difference using total N and inorganic N values, and soil texture was determined

using the Bouyoucos hydrometer technique (Bouyoucos 1962). KCl extractions were analyzed

for [NH4+] and [NO3

-] using colorimetric methods on an Astoria Pacific II continuous flow

autoanalyzer.

Resin bags were constructed using 20 g of mixed bed ion exchange resin (J.T Baker

#JT4631-1) tied inside a piece of un-dyed nylon stocking material (Binkley et al. 1992). One

resin bag was incubated at 10 cm soil depth in each microplot (N = 10 / plot). Sampling occurred

in four summer periods (June-August in 2007 and 2008, June-October in 2009, and June-

September in 2010) and three winter periods (August-June in 2007-8 and 2008-9, and October-

June in 2009-10). Beetle + salvage plots were not measured during the salvage period in summer

84

2009. Upon retrieval, resins were removed from the nylon and extracted in 2M KCl and analyzed

in the same manner as the soil samples described above.


To test for similarity in pre-salvage site conditions between the beetle only and beetle +

salvage stands, we performed paired t tests (α < 0.05) on topographic (elevation, slope, aspect),

soil (texture, N, organic matter (OM), pH, and cations), vegetation (basal area, density of <1.4m

stems, and ground cover), and litter (quantity and quality) metrics measured in 2007. Potential

effects of post-beetle salvage logging on these variables were tested by calculating plot-level

changes between 2007 and 2010, and performing paired t-tests (α < 0.05) of this change in beetle

only vs. beetle + salvage plots. Soil temperatures were analyzed using mid-summer (July 1-

August 31) and mid-winter (January 1-February 28) data from before and after the salvage

period. For each soil depth, we first calculated the plot-level difference in mean daily

temperatures between like calendar days before and after salvage, then performed paired t-tests

for a salvage effect on these differences (N = 62 for summer; N = 59 for winter). Resin bag data

within each sampling period were calculated as the rate of N accumulation (µg N g resin-1 day-1)

and the mass of N collected (µg N g resin-1). Potential effects of post-beetle salvage logging on

resin bag N were tested using paired t-tests of the change over time in beetle only vs. beetle +

salvage plots using only the years immediately before and after salvage (2010-2008). All

variables were tested for normality and transformed when necessary to meet the assumptions of

statistical methods. Unless otherwise noted, reported variance measures are two standard errors.

RESULTS

Pre-salvage site conditions.

85

Beetle only and beetle + salvage plots were similar (P > 0.05) in all topographic,

vegetation, litter, cover, and soil metrics except for organic soil depth, which was deeper by 0.5

cm in beetle-only sites (Table 1). Mean elevation of all plots was 2518 ± 13 m, with a mean

slope of 14.3 ± 4º. Lodgepole pine averaged 97% of the total basal area (live + dead) of all

stands, and beetle mortality averaged 60% of lodgepole basal area (Table 1). The density of all

saplings was highly variable among all plots prior to salvage but was not significantly different

between beetle only and beetle + salvage plots (Table 1). Density of lodgepole pine saplings also

did not differ between beetle and beetle + salvage plots (Table 1), and was 61% and 62% of the

total density in each, respectively.

Vegetation and ground cover

As expected, salvage logging reduced total lodgepole pine basal area by 90%, which

included small amounts of both live and non-beetle killed dead basal area in addition to beetle-

killed trees (Figure 2a). Live unattacked basal area declined similarly in both beetle only and

beetle + salvage sites, and beetle-killed basal area increased in beetle-only plots (Figure 2a). In

the understory, salvage reduced total biotic cover, largely in response to a decline in forb cover,

and there was no effect on graminoid or shrub cover (Figure 2b). Ground cover of bare soil

increased after salvage from 0.1 to 8.4%, and coarse wood cover increased from 2.3 to 11.6 %

(Figure 2b).

Salvage logging had a discernable effect on sapling density despite high variability in

density change (both positive and negative) between 2007 and 2010 in both beetle only and

beetle + salvage plots (P = 0.045). Total sapling density in beetle + salvage plots declined on

average 1892 ± 1648 stems ha-1 down to a mean of 1675 ± 1001 stems ha-1. However, though the

densities of both 0-30cm and 30-140cm size classes declined in beetle + salvage sites, when

86

compared to changes in beetle only plots an effect of salvage was limited to the 30-140cm size

class (Figure 2a). For lodgepole pine saplings, there was no effect of salvage on either the total

density or the density of either the 0-30 or 30-140 cm size classes (Figure 3b).

Soil microclimate

Salvage logging had a warming effect on both mid-summer and mid-winter soil

temperatures at both the litter-soil interface and 10cm depth. Salvage increased mid-summer

litter-soil interface temperature by 0.8 ± 0.2 ºC, and 10 cm depth temperature by 1.1 ± 0.1 ºC (P

< 0.001 for both; Figure 3a). Mid-winter temperatures were 0.2 ± 0.05 and 0.3 ± 0.04 ºC warmer

at the litter-soil interface and 10 cm depth, respectively (P < 0.001 for both; Figure 3b). Mean

daily temperatures at the litter-soil interface peaked at 17 ºC in each of the three sampled

summers, and reached minimums of -3 and -9 ºC in the two sampled winters. At 10 cm depth,

temperatures peaked at 12-14 ºC in summers, and reached lows of 3 and 6 ºC in winters.


Salvage had little effect on litter amounts or N content, though several litter variables

changed over time across all plots. Litter mass increased by 37% (1600 to 2200 g m-2) between

2007 and 2010 but was unaffected by salvage (Figure 5a). Litter depth averaged 2.7 ± 0.2 cm and

did not change over time or with salvage (Figure 5b). Fresh (<1yr) needle litter %N was 0.73 ±

0.03 % across all sites and did not change with either time or salvage (Figure 5d). Though the

%N of all needle litter was unchanged in beetle-only sites, it increased following salvage (Figure

5e) from 0.72 ± 0.08 to 0.94 ±0.07% but was insufficient to cause a significant salvage effect in

total litter %N (Figure 5f). Total litter %N did, however, increase across all plots from 0.81 ±

0.06 to 0.98 ± 0.06% between 2007 and 2010 (Figure 5f). Increased litter mass and increased

87

total litter %N both contributed to a 50% increase over time in the litter N pool across all plots

from 1473 ± 148 to 2177 ± 295 g N m-2, with no additional effect of salvage (Figure 5c).

Soil and resin bag N

Pools of extractable soil inorganic N changed through time and with salvage. Soil NH4+

tripled across all plots from 5.7 ± 0.9 to 17.3 ± 0.9 µg N g soil-1, with no additional effect of

salvage (Figure 6a). In contrast, soil NO3- was unchanged in beetle-only plots through time but

almost tripled in beetle + salvage plots from 0.6 ± 0.2 to 1.6 ± 0.5 µg N g soil-1 from 2007 to

2010 (Figure 6b). However, this difference in soil NO3- was insufficient to cause an effect of

salvage on total extractable inorganic N, which was dominated by NH4+ and increased from 6.2 ±

1.0 to 18.9 ± 3.0 µg N g soil-1 across all plots between 2007 and 2010.

Resin bag NH4+ accumulation rate increased throughout the study in all plots (Figure 7a).

In the pre-salvage period, NH4+ accumulation rate increased in all plots from < 0.10 to ~0.20 µg

N g resin-1 day-1 between summer 2007 and summer 2008. NH4+ accumulation rate continued to

increase through the post-salvage period in all plots, reaching ~0.35 µg N g resin-1 day-1 in the

summer of 2010. Patterns of NH4+ mass collected paralleled those of accumulation rate (Figure

7b). There was no effect of salvage on the changes in seasonal or annual NH4+ accumulation

rates, or seasonal mass collected between the year before and the year after the salvage period (P

> 0.05). Annual mass of NH4+ collected was also unaffected by salvage (Figure 8). Within the

consistently increasing trend, resin bag NH4+ accumulation oscillated between higher rates in

summer and lower rates in winter (Figure 7a). Resin bag NO3- accumulation rate was consistent

throughout the study and almost always under 0.10 µg N g resin-1 day-1 during both summer and

winter (Figure 7b), with parallel patterns in total mass collected (Figure 7d). There was no effect

88

of salvage on seasonal or annual NO3-accumulation rates or seasonal mass of NO3

- collected (P >

0.05). Annual mass of NO3-collected was also unaffected by salvage (Figure 8).

DISCUSSION

In this study we tested the effects of post-disturbance salvage logging on the structure,

microclimate, and N cycling of bark beetle-killed lodgepole pine forests. Bark beetle disturbance

alone redistributes material among the canopy and forest floor, and influences both the

microclimate and N cycling of the litter-soil profile compared to undisturbed forests (Griffin et

al. 2011). We hypothesized that salvage logging of beetle-killed forests would cause further

canopy disturbance and litter inputs, as well as new disturbance to saplings, understory

vegetation, and the forest floor which together may in turn alter patterns of N cycling in the

litter-soil profile compared to beetle-kill alone. Salvage had the greatest impact on forest

structure, though there were substantive effects on ground cover, soil microclimate and litter and

soil N cycling as well. In addition to declines in total basal area, density of 30 to 140-cm high

saplings, and % cover of forbs and all plants, salvage caused an increase in mid-summer and

mid-winter soil temperatures, %N content in needle litter, extractable soil NO3-, and the percent

cover of coarse wood.

Tree sapling density and regeneration potential

Salvage-induced declines in the total sapling density observed here are consistent with

those found in other salvaged forests (Jonasova and Prach 2004, Keyser et al. 2009). However,

salvage did not affect the smallest size classes (0 to 30-cm high) of all saplings, or any size class

of lodgepole pine sapling. This may be due to the undisturbed forest floor in post-beetle forests

which may protect smaller stems from trampling or erosion compared to post-fire forests, and is

89

a common result following salvage (Vyse et al. 2009). The average density of stems <1.4m high

in salvaged plots of this study remained above 1600 stems ha-1, which is considered sufficient for

stand replacement (Vyse et al. 2009), and also remained dominated by lodgepole pine (>50% of

total density). Effects of salvage on sapling density are variable among forest types, pre-salvage

disturbance types, and salvage method. In a study of post-southern pine beetle (Dendroctonus

frontalis Zimmermann) salvage logging in loblolly pine (Pinus taeda L.), Coleman et al. (2008)

concluded that cut and leave treatments did not change the successional trajectory of beetle-

killed forests. Similarly, Peterson and Leach (2008) found no effect of post-windthrow salvage

on sapling densities in loblolly pine, though there was an effect on species composition. In

contrast, post-beetle (Ips typographus L.) clear-cut salvage logging in a Norway spruce (Picea

abies (L.) Karst.) forest reduced the density of spruce saplings and promoted early successional

understory species (Jonasova and Prach 2004, 2008).

Microclimate and ground cover

The salvage effect of 1 ºC increase in temperature below the litter and at 10-cm depth

observed here following beetle salvage is consistent in magnitude with remotely-sensed surface

temperatures in beetle-killed spruce and clear-cut forests of the Czech Republic, which showed a

difference of 2 ºC between beetle-killed and clear-cut forest. (Hais and Kucera 2008). However,

initial beetle disturbance has been shown to lower soil temperature compared to undisturbed

forests due to increased litter input (Byers 1984, Griffin et al. 2011). Litter mass did not increase

with salvage in this study and thus likely did not provide any additional cooling effect on soils.

The removal of tree boles and the decreases in biotic ground cover and in the density of stems

<1.4 m high may have reduced soil shading, and could explain how salvage increased soil

90

temperatures. Warmer temperatures may be associated with drier soils, which would have a

suppressing effect on soil N turnover.

Increased cover of coarse wood after salvage logging is likely to create more variability

in forest floor microsite conditions, influencing the rates of litter decomposition and subsequent

return of inorganic N to soils at fine spatial scales (Remsburg and Turner 2006). However,

similar changes are likely to occur in beetle-kill only forests as dead trees begin to fall, and over

time total coarse wood loading in beetle-kill only forests will likely be much higher than in

salvaged forests (Lewis 2009). Decreases in total biotic cover observed 1 year following salvage

are not likely to remain (Kurulok and Macdonald 2007), as a flush of understory growth spurred

by increased water, light, and nutrients is often observed following both beetle mortality (Stone

and Wolfe 1996) and post-beetle salvage harvest (Jonasova and Prach 2008). Though highly

variable and not significant in paired analyses, percent cover of bare soil showed an increasing

trend in salvaged sites and possibly contributed to significant decreases in biotic cover. However,

over time these areas are likely to be colonized by understory vegetation. With more mineral soil

substrate exposed as seedbed, colonization by pioneer species may alter understory community

composition in salvaged sites relative to beetle-kill alone. In a Colorado spruce-fir forest salvage

logged after windthrow, understory communities shifted to dominance by graminoids (del Rio

2006). Thus, despite initial differences in biotic cover after 1 year, prediction of long term

impacts of salvage on understory composition is difficult.

Litter and Soils

Several metrics of litter and soil N cycling changed similarly over time in all plots,

consistent with expectations of canopy deterioration beetle-killed forests (Griffin et al. 2011).

Total litter mass, %N, and litter N pool size were unaffected by salvage logging, but all increased

91

over the three years between sampling. These results suggest that either (1) most foliage had

fallen from beetle-killed trees in all plots prior to salvage, or (2) that the amount of salvage-

induced litterfall in beetle + salvage plots was equivalent to litterfall in beetle-kill only plots over

the same time period. In either scenario, salvage does not appear to increase the amount of litter

mass or N input which would already occur following beetle outbreak. Salvage did increase the

%N of all needle litter, although this was insufficient to influence the total litter N pool size.

This may be due to salvage-induced inputs of green needle litter from trees that survived the

outbreak and had elevated foliar N content (Griffin et al. 2011).

Similarly, patterns of change in soil and resin bag N of salvaged forests were mostly

consistent with those observed in beetle-killed only forests. Nitrogen availability is largely

determined by the balance between plant uptake and inputs of organic N to the forest floor.

Equivalent live basal areas between beetle only and beetle + salvage plots over time suggest that

canopy plant demand for N was not affected by salvage. Inputs of litter N suggest N supply was

also similar, which together may explain why few differences in soil N were observed.

Extractable soil NO3- pool was the only soil N metric that differed over time between beetle-only

and beetle + salvage plots, and though still low, approximately tripled in beetle + salvage plots

compared to beetle-kill alone. Increased soil NO3- could be due to reduced cover of forbs, which

preferentially take up NO3- over NH4

+ (Miller and Bowman 2002, Falkengren-Grerup et al.

2004).

An increase in NO3- over time was not detected with buried resin bags in either beetle

only or beetle + salvage plots, while resin bag NH4+ increased equally in both. With the

exception of increased (but still low) soil NO3-, salvage does not appear to substantially alter soil

N dynamics relative to the changes that would already occur in beetle-killed forests. Other

92

studies of tree mortality impacts on soil NO3- in Rocky Mountain conifers are consistent with

these results, and show that NO3- production and export remains low even after substantial tree

removal or death (Knight et al. 1991, Parsons et al. 1994, Prescott et al. 2003, Thiel and Perakis

2009). Beetle and salvage induced inputs of litter have high C:N ratios and may serve as N sinks

in the early stages of decomposition (Fahey et al. 1985, Remsburg and Turner 2006) and limit

the accumulation of N in soils.

CONCLUSIONS

Post-beetle salvage logging had detectable influences on ground cover, soil microclimate,

and N cycling through the litter-soil profile during the initial year following tree harvest.

However, the effects of salvage harvest were of less magnitude than we expected. Although

sapling density declined by 50%, the density of lodgepole pine saplings did not change and

appeared adequate for stand replacement. Salvage increased soil temperatures by ~ 1 ºC at both

the litter-soil interface and at 10cm depth, and increased the %N in all-needle litter but not in the

total litter N pool. With no effect on plant N demand (as inferred from basal area) or on litter N

inputs, salvage also did not have a substantial effect on soil N other than small increases in

extractable soil NO3-. Rather, N cycling in the litter and soils of both beetle only and beetle +

salvage sites changed over time in a manner consistent with that expected following bark beetle

outbreak with few modifications by salvage harvest. These data suggest that the harvest of

beetle-killed trees for wood products can have small initial consequences for many ecosystem

response variables. However, loss of snags and the future inputs of coarse wood to the forest

floor are likely to be pronounced over longer time frames.

93

ACKNOWLEDGEMENTS

We thank Emily Stanley, James Thoyre, and the Center for Limnology at UW for access

to laboratory equipment. For help in the field and the laboratory, we appreciate the assistance of

many University of Wisconsin students: Heather Lumpkin, Erin Mellenthin, Alex Rahmlow,

Amanda Rudie, Ben Ruh, Corey Olsen, Ryan Peasley, Greg Skupien, Lucille Marescot, Jaclyn

Entringer, Emily Ruebl, Ashton Mieritz, Shane Armstrong, and Nick Labonte.

REFERENCES

Astrup, R., K. D. Coates, and E. Hall. 2008. Recruitment limitation in forests: Lessons from an unprecedented mountain pine beetle epidemic. Forest Ecology and Management 256:1743-1750. Bentz, B. J., S. Kegley, K. Gibson, and R. Thier. 2005. A test of high-dose verbenone for stand- level protection of lodgepole and whitebark pine from mountain pine beetle (Coleoptera : Curculionidae : Scolytinae) attacks. Journal of Economic Entomology 98:1614-1621. Binkley, D., R. Bell, and P. Sollins. 1992. Comparison of Methods for Estimating Soil-Nitrogen Transformations in Adjacent Conifer and Alder-Conifer Forests. Canadian Journal of Forest Research 22:858-863. Blouin, V. M., M. G. Schmidt, C. E. Bulmer, and M. Krzic. 2005. Mechanical disturbance impacts on soil properties and lodgepole pine growth in British Columbia's central interior. Canadian Journal of Soil Science 85:681-691. Blouin, V. M., M. G. Schmidt, C. E. Bulmer, and M. Krzic. 2008. Effects of compaction and water content on lodgepole pine seedling growth. Forest Ecology and Management 255:2444-2452. Bouyoucos, G. J. 1962. Hydrometer method improved for making particle and size analysis of soils. Agronomy Journal 54:464-465. Bulmer, C. E., and D. G. Simpson. 2005. Soil compaction and water content as factors affecting the growth of lodgepole pine seedlings on sandy clay loam soil. Canadian Journal of Soil Science 85:667-679. Byers, J. A. 1984. Electronic Multiprobe Thermometer and Multiplexer for Recording Temperatures of Microenvironments in the Forest Litter Habitat of Bark Beetles (Coleoptera, Scolytidae). Environmental Entomology 13:863-867.

94

Chatterjee, A., G. F. Vance, E. Pendall, and P. D. Stahl. 2008. Timber harvesting alters soil carbon mineralization and microbial community structure in coniferous forests. Soil Biology & Biochemistry 40:1901-1907. Choi, W. J., S. X. Chang, M. P. Curran, H. M. Ro, M. Kamaluddin, and J. J. Zwiazek. 2005. Foliar delta C-13 and delta N-15 response of lodgepole pine and Douglas-fir seedlings to soil compaction and forest floor removal. Forest Science 51:546-555. Coleman, T. W., S. R. Clarke, J. R. Meeker, and L. K. Rieske. 2008. Forest composition following overstory mortality from southern pine beetle and associated treatments. Canadian Journal of Forest Research 38:1406-1418. del Rio, C. M. R. 2006. Changes in understory composition following catastrophic windthrow and salvage logging in a subalpine forest ecosystem. Canadian Journal of Forest Research 36:2943-2954. Despain, D. G. 1990. Yellowstone Vegetation. Roberts Rinehart, Boulder. Donato, D. C., J. B. Fontaine, J. L. Campbell, W. D. Robinson, J. B. Kauffman, and B. E. Law. 2006. Post-wildfire logging hinders regeneration and increases fire risk. Science 311:352- 352. Edmonds, R. L., and A. Eglitis. 1989. The Role of the Douglas-Fir Beetle and Wood Borers in the Decomposition of and Nutrient Release from Douglas-Fir Logs. Canadian Journal of Forest Research 19:853-859. Fahey, T. J., J. B. Yavitt, J. A. Pearson, and D. H. Knight. 1985. The Nitrogen-Cycle in Lodgepole Pine Forests, Southeastern Wyoming. Biogeochemistry 1:257-275. Falkengren-Grerup, U., A. Michelsen, M. O. Olsson, C. Quarmby, and D. Sleep. 2004. Plant nitrate use in deciduous woodland: the relationship between leaf N, N-15 natural abundance of forbs and soil N mineralisation. Soil Biology & Biochemistry 36:1885- 1891. Fettig, C. J., K. D. Klepzig, R. F. Billings, A. S. Munson, T. E. Nebeker, J. F. Negron, and J. T. Nowak. 2007. The effectiveness of vegetation management practices for prevention and control of bark beetle infestations in coniferous forests of the western and southern United States. Forest Ecology and Management 238:24-53. Flint, C. G. 2006. Community perspectives on spruce beetle impacts on the Kenai Peninsula, Alaska. Forest Ecology and Management 227:207-218. Flint, C. G., and R. Haynes. 2006. Managing forest disturbances and community responses: Lessons from the Kenai Peninsula, Alaska. Journal of Forestry 104:269-275.

95

Gillette, N. E., J. D. Stein, D. A. Owen, J. N. Webster, G. O. Fiddler, S. R. Mori, and D. L. Wood. 2006. Verbenone-releasing flakes protect individual Pinus contorta trees from attack by Dendroctonus ponderosae and Dendroctonus valens (Coleoptera : Curculionidae, Scolytinae). Agricultural and Forest Entomology 8:243-251. Goodman, L. F., and B. A. Hungate. 2006. Managing forests infested by spruce beetles in south- central Alaska: Effects on nitrogen availability, understory biomass, and spruce regeneration. Forest Ecology and Management 227:267-274. Griffin, J. M., M. G. Turner, and M. Simard. 2011. Nitrogen cycling following mountain pine beetle disturbance in lodgepole pine forests of Greater Yellowstone. Forest Ecology and Management 261:1077-1089. Hais, M., and T. Kucera. 2008. Surface temperature change of spruce forest as a result of bark beetle attack: remote sensing and GIS approach. European Journal of Forest Research 127:327-336. Hanson, J. J., and J. D. Stuart. 2005. Vegetation responses to natural and salvage logged fire edges in Douglas-fir/hardwood forests. Forest Ecology and Management 214:266-278. Herrmann, S., and C. E. Prescott. 2008. Mass loss and nutrient dynamics of coarse woody debris in three Rocky Mountain coniferous forests: 21 year results. Canadian Journal of Forest Research 38:125-132. Jeans-Williams, N. L., and J. H. Borden. 2006. Evaluation of two pheromone baits for containment and concentration of attack by the western balsam bark beetle, Dryocoetes confusus (Coleoptera : Curculionidae). Western Journal of Applied Forestry 21:27-32. Jonasova, M., and K. Prach. 2004. Central-European mountain spruce (Picea abies (L.) Karst.) forests: regeneration of tree species after a bark beetle outbreak. Ecological Engineering 23:15-27. Jonasova, M., and K. Prach. 2008. The influence of bark beetles outbreak vs. salvage logging on ground layer vegetation in Central European mountain spruce forests. Biological Conservation 141:1525-1535. Kamaluddin, M., S. X. Chang, M. P. Curran, and J. J. Zwiazek. 2005. Soil compaction and forest floor removal affect early growth and physiology of lodgepole pine and Douglas-fir in British Columbia. Forest Science 51:513-521. Keyser, T. L., F. W. Smith, and W. D. Shepperd. 2009. Short-term impact of post-fire salvage logging on regeneration, hazardous fuel accumulation, and understorey development in ponderosa pine forests of the Black Hills, SD, USA. International Journal of Wildland Fire 18:451-458.

96

Klutsch, J. G., J. F. Negron, S. L. Costello, C. C. Rhoades, D. R. West, J. Popp, and R. Caissie. 2009. Stand characteristics and downed woody debris accumulations associated with a mountain pine beetle (Dendroctonus ponderosae Hopkins) outbreak in Colorado. Forest Ecology and Management 258:641-649. Knight, D. H., J. B. Yavitt, and G. D. Joyce. 1991. Water and Nitrogen Outflow from Lodgepole Pine Forest after 2 Levels of Tree Mortality. Forest Ecology and Management 46:215- 225. Kranabetter, J. M., P. Sanborn, B. K. Chapman, and S. Dube. 2006. The contrasting response to soil disturbance between lodgepole pine and hybrid white spruce in subboreal forests. Soil Science Society of America Journal 70:1591-1599. Kurulok, S. E., and S. E. Macdonald. 2007. Impacts of postfire salvage logging on understory plant communities of the boreal mixedwood forest 2 and 34 years after disturbance. Canadian Journal of Forest Research 37:2637-2651. Laiho, R., and C. E. Prescott. 1999. The contribution of coarse woody debris to carbon, nitrogen, and phosphorus cycles in three Rocky Mountain coniferous forests. Canadian Journal of Forest Research 29:1592-1603. Laiho, R., and C. E. Prescott. 2004. Decay and nutrient dynamics of coarse woody debris in northern coniferous forests: a synthesis. Canadian Journal of Forest Research 34:763-777. Lewis, D. 2009. Stand and landscape-level simulations of mountain pine beetle (Dendroctonus ponderosae) and salvage logging effects on live tree and deadwood habitats in south- central British Columbia, Canada. Forest Ecology and Management 258:S24-S35. Lopushinsky, W., D. Zabowski, and T. D. Anderson. 1992. Early Survival and Height Growth of Douglas-Fir and Lodgepole Pine-Seedlings and Variations in Site Factors Following Treatment of Logging Residues. Usda Forest Service Pacific Northwest Research Station Research Paper:U1-22. McCambridge, W. F., M. J. Morris, and C. B. Edminster. 1982. Herbage production under ponderosa pine killed by the mountain pine beetle in Colorado. Usda Forest Service Intermountain Research Station Research Paper RM-416. McFarlane, B. L., and D. O. T. Witson. 2008. Perceptions of ecological risk associated with mountain pine beetle (Dendroctonus ponderosae) infestations in Banff and Kootenay National Parks of Canada. Risk Analysis 28:203-212. McMillin, J. D., and K. K. Allen. 2000. Impacts of Douglas-fir beetle on overstory and understory conditions of Douglas-fir stands, Shoshone National Forest, Wyoming. Technical Report R2-64, USDA Forest Service, Golden, CO.

97

Miller, A. E., and W. D. Bowman. 2002. Variation in nitrogen-15 natural abundance and nitrogen uptake traits among co-occurring alpine species: do species partition by nitrogen form? Oecologia 130:609-616. Morehouse, K., T. Johns, J. Kaye, and A. Kaye. 2008. Carbon and nitrogen cycling immediately following bark beetle outbreaks in southwestern ponderosa pine forests. Forest Ecology and Management 255:2698-2708. Nigh, G. D., J. A. Antos, and R. Parish. 2008. Density and distribution of advance regeneration in mountain pine beetle killed lodgepole pine stands of the Montane Spruce zone of southern British Columbia. Canadian Journal of Forest Research 38:2826-2836. Paine, R. T., M. J. Tegner, and E. A. Johnson. 1998. Compounded perturbations yield ecological surprises. Ecosystems 1:535-545. Parsons, W. F. J., D. H. Knight, and S. L. Miller. 1994. Root Gap Dynamics in Lodgepole Pine Forest : Nitrogen Transformations in Gaps of Different Size. Ecological Applications 4:354-362. Peterson, C. J., and A. D. Leach. 2008. Limited salvage logging effects on forest regeneration after moderate-severity windthrow. Ecological Applications 18:407-420. Prescott, C. E. 2002. The influence of the forest canopy on nutrient cycling. Tree Physiology 22:1193-1200. Prescott, C. E., G. D. Hope, and L. L. Blevins. 2003. Effect of gap size on litter decomposition and soil nitrate concentrations in a high-elevation spruce-fir forest. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 33:2210-2220. Progar, R. A. 2005. Five-year operational trial of verbenone to deter mountain pine beetle (Dendroctonus ponderosae; Coleoptera : Scolytidae) attack of lodgepole pine (Pinus contorta). Environmental Entomology 34:1402-1407. Raffa, K. F., B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H. Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic amplification: The dynamics of bark beetle eruptions. Bioscience 58:501-517. Remsburg, A. J., and M. G. Turner. 2006. Amount, position, and age of coarse wood influence litter decomposition in postfire Pinus contorta stands. Canadian Journal of Forest Research 36:2112-2123. Rocca, M. E., and W. H. Romme. 2009. Beetle-infested forests are not "destroyed". Frontiers in Ecology and the Environment 7:71-72. Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine beetle in lodgepole pine forests. Pages 3-66 in L. Safranyik and B. Wilson, editors. The

98

mountain pine beetle: A synthesis of biology, management, and impacts on lodgepole pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, BC. Safranyik, L., T. L. Shore, and D. A. Linton. 1999. Attack by bark beetles (Coleoptera : Scolytidae) following spacing of mature lodgepole fine (Pinaceae) stands. Canadian Entomologist 131:671-685. Shore, T. L., L. Safranyik, B. C. Hawkes, and S. W. Taylor. 2006. Effects of the mountain pine beetle on lodgepole pine stand structure and dynamics. Pages 95-114 in L. Safranyik and B. Wilson, editors. The mountain pine beetle: A synthesis of biology, management, and impacts on lodgepole pine. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, B.C. Smethurst, P. J., and E. K. S. Nambiar. 1990. Effects of Slash and Litter Management on Fluxes of Nitrogen and Tree Growth in a Young Pinus-Radiata Plantation. Canadian Journal of Forest Research 20:1498-1507. Stone, W. E., and M. L. Wolfe. 1996. Response of understory vegetation to variable tree mortality following a mountain pine beetle epidemic in lodgepole pine stands in northern Utah. Vegetatio 122:1-12. Thiel, A. L., and S. S. Perakis. 2009. Nitrogen dynamics across silvicultural canopy gaps in young forests of western Oregon. Forest Ecology and Management 258:273-287. Underwood, A. J. 1994. On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecological Applications 4:3-15. UWSPAL. 2010. Soil and Plant Analysis.in J. Peters, editor. Wisconsin Procedures for Soil Testing, Plant Analysis and Feed & Forage Analysis. University of Wisconsin Soil and Plant Analysis Lab, Madison, WI.http://uwlab.soils.wisc.edu/procedures.htm. Vyse, A., C. Ferguson, D. J. Huggard, J. Roach, and B. Zimonick. 2009. Regeneration beneath lodgepole pine dominated stands attacked or threatened by the mountain pine beetle in the south central Interior, British Columbia. Forest Ecology and Management 258:S36- S43. WRCC. 2010. Cora, WY, climate summary Western Regional Climate Center. http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?wy2054.

99

TABLES

Table 1. Pre-salvage characteristics for beetle-killed only and beetle-killed plus salvage-logged

sites. N = 8 sites per treatment. Boldface indicates significant paired t test at α = 0.05.

Disturbance type Site characteristic Beetle only Beetle + Salvage P Topography Elevation (m) 2510 ± 22 2526 ± 12 0.0899 Aspect (SW index) a -0.7 ± 0.4 -0.1 ± 0.6 0.2250 Slope (º) 14 ± 6 15 ± 5 0.7851 Soils Texture & Structure Sand (%) 61 ± 5 57 ± 5 0.3128 Silt (%) 26 ± 4 29 ± 3 0.1969 Clay (%) 13 ± 2 14 ± 2 0.8483 Bulk density (g cm-3) 0.69 ± 0.06 0.72 ± 0.07 0.5277 Organic soil depth (cm) 3.2 ± 0.3 2.7 ± 0.2 0.0137 Organic matter (%) 4.3 ± 0.8 4.7 ± 1.0 0.5424 Cations pH 5.4 ± 0.1 5.3 ± 0.2 0.4905 Ca (µg g-1; exch.) 1132 ± 293 1251 ± 361 0.5691 Mg (µg g-1; exch.) 131 ± 24 141 ± 34 0.5415 K (µg g-1; exch.) 187 ± 37 185 ± 25 0.8680 P (µg g-1; exch.) 25 ± 8 28 ± 5 0.4172 Nitrogen NH4

+ (µg N g-1) 5.3 ± 1.4 6.0 ± 1.1 0.3769 NO3

- (µg N g-1) 0.5 ± 0.3 0.6 ± 0.3 0.4116 Organic N (mg N g-1) 0.86 ± 0.25 0.94 ± 0.28 0.6978 Litter Mass (g m-2) 1614 ± 236 1668 ± 160 0.7698 Depth (cm) 2.7 ± 0.4 2.3 ± 0.4 0.1913 N pool size (g N m-2) 12.6 ± 2.2 13.9 ± 1.8 0.3794 <1 yr old needle %N 0.73 ± 0.07 0.69 ± 0.05 0.3402 All needles %N 0.74 ± 0.07 0.72 ± 0.08 0.6538 Total litter %N 0.78 ± 0.06 0.84 ± 0.10 0.303 P. contorta basal area (m2 ha-1) Live unattacked 8.3 ± 2.8 5.6 ± 2.8 0.0978 Live attacked 2.1 ± 2.2 0.6 ± 1.2 0.3050 Dead unattacked 5.0 ± 3.1 8.0 ± 4.3 0.2290 Dead attacked 22.4 ± 4.1 21.9 ± 5.0 0.8156 Sapling density (stems ha-1) All species 7133 ± 3380 3567 ± 1855 0.1010b

P. contorta 4383 ± 2854 2225 ± 1326 0.2798b

Ground cover (%) Bare soil 0.4 ± 0.5 0.1 ± 0.1 0.1705 Litter 42 ± 9 41 ± 10 0.7418 Coarse wood 2.6 ± 3.2 2.3 ± 1.7 0.8202 Total biotic 62 ± 11 60 ± 13 0.7322 Graminoid 16 ± 6 7 ± 4 0.0607 Shrub 7 ± 6 3 ± 3 0.0976 Forb 37 ± 11 49 ± 9 0.0664 a calculated as: cos(aspect-225)

b paired t test performed on log10 transformed data

100

FIGURE LEGENDS

Figure 1. Study area, Green River Lakes Region of the Bridger-Teton National Forest.

Eight plot pairs were established in summer 2007, with one member inside a polygon designated

for salvage harvest and the other outside. All forests in the field of view are dominated by

lodgepole pine, and mountain pine beetle activity was extensive throughout the area at the time

of initial sampling (2007). Salvage harvests occurred in the summer of 2009, and post-salvage

data were collected in summer 2010.

Figure 2. Stand structure and ground cover change in beetle only and beetle + salvage

plots following logging. N = 8 pairs of plots. Error bars = 2 standard error. (A) Live and dead

basal area change (2010 – 2007) in attacked and unattacked lodgepole pine; (B) Ground cover

change in abiotic (bare soil, litter, and coarse wood) and biotic (total and by plant functional

group) classes. An asterisk indicates a significant difference at α = 0.05.

Figure 3. Change in sapling density in beetle only and beetle + salvage plots following

logging. N = 8 pairs of plots. Error bars = 2 standard error. (A) all saplings; (B) lodgepole pine

saplings. An asterisk indicates a significant difference at α = 0.05.

Figure 4. Average daily soil temperature in beetle only and beetle + salvage plots. Beetle +

salvage sites were not instrumented during the salvage period in summer 2009. Variable width of

each plotted time series = daily 95% confidence interval. Beetle-only data (black) is overlaid on

beetle + salvage data (gray). (A) temperature at the litter-organic soil interface; (B) temperature

at 10-cm soil depth. Salvage logging increased both mid-summer (July-August) temperature at

101

both soil depths by approximately 1 ºC, and mid-winter (January-February) temperatures by 0.2

ºC (P < 0.001; time periods used in analyses are shown in boxes).

Figure 5. Change in litter quantity and quality in beetle only and beetle + salvage plots

following logging. N = 8 pairs of plots. Error bars = 2 standard error. (A) litter mass; (B) litter

depth; (C) %N in <1yr old needle litter; (D) %N in all-needle litter; (E) %N in total litter; (F)

total litter N pool size. An asterisk indicates a significant difference at α = 0.05.

Figure 6. Change in soil inorganic N pool sizes in beetle only and beetle + salvage plots

following logging. N = 8 pairs of plots. Error bars = 2 standard error. (A) extractable NH4+; (B)

extractable NO3-; (C) total extractable inorganic N (TIN). An asterisk indicates a significant

difference at α = 0.05.

Figure 7. Resin bag N in beetle only and beetle + salvage plots. N = 8. Error bars = 2 standard

error. Beetle + salvage sites were not measured during the salvage period in summer of 2009. (A)

Resin bag NH4+ accumulation rate; (B) Resin bag NO3

- accumulation rate; (C) Mass of resin bag

NH4+; (D) Mass of resin bag NO3

-.

Figure 8. Change in annual mass of resin bag N in beetle only and beetle + salvage plots

following logging. N = 8 pairs of plots. Error bars = 2 standard error. (A) NH4+; (B) NO3

-; (C)

total inorganic N (TIN).

102

FIGURES

Figure 1

103

Figure 2

-28

-24

-20

-16

-12

-8

-4

0

4

8

Liveunattacked

Liveattacked

Deadunattacked

Deadattacked

Cha

nge

in b

asal

are

a (m

2 ha-1

)

Beetle onlyBeetle + Salvage

*

*

A)

-50

-40

-30

-20

-10

0

10

20

30

40

Bare soil Litter Coarsewood

Totalbiotic

Graminoid Shrub Forb

Cha

nge

in p

erce

nt c

over

(%)

*

* *

B)

104

Figure 3

-4000

-3000

-2000

-1000

0

1000

2000

3000

4000

Cha

nge

in to

tal d

ensi

ty ..

(sap

lings

ha-1

)


A)

**

-4000

-3000

-2000

-1000

0

1000

2000

3000

4000

All stems<140 cm

0-30 cm 30-140 cm

Cha

nge

in lo

dgep

ole

pine

..de

nsity

(sap

lings

ha-1

) .

B)

105

Figure 4

-8.0-6.0-4.0-2.00.02.04.06.08.0

10.012.014.0

6/11

/200

87/

11/2

008

8/11

/200

89/

11/2

008

10/1

1/20

0811

/11/

2008

12/1

1/20

081/

11/2

009

2/11

/200

93/

11/2

009

4/11

/200

95/

11/2

009

6/11

/200

97/

11/2

009

8/11

/200

99/

11/2

009

10/1

1/20

0911

/11/

2009

12/1

1/20

091/

11/2

010

2/11

/201

03/

11/2

010

4/11

/201

05/

11/2

010

6/11

/201

07/

11/2

010

8/11

/201

09/

11/2

010

10 c

m d

epth

tem

p (C

) .

Beetle + Salvage

Beetle only

B)

-10.0-8.0-6.0-4.0-2.00.02.04.06.08.0

10.012.014.016.018.0

Litte

r-so

il in

terfa

ce te

mp

(C)

.

A)Salvage periodPre-salvage Post-salvage

106

Figure 5

0

200

400

600

800

1000

1200C

hang

e in

litte

r mas

s (g

m-2

) .

A)

-0.5

0.0

0.5

1.0

1.5

Cha

nge

in li

tter d

epth

(cm

) .

B)

0

2

4

6

8

10

12

14

Cha

nge

in li

tter N

poo

l (g

N m-2

).

C)

-0.1

0.0

0.1

0.2

0.3

0.4

Cha

nge

in <

1 yr

old

nee

dle

litte

r %N

.D)

-0.1

0.0

0.1

0.2

0.3

0.4

Cha

nge

in a

ll ne

edle

litte

r %N

.

E)

*

-0.1

0.0

0.1

0.2

0.3

0.4

Cha

nge

in to

tal l

itter

%N

.

F)

107

Figure 6

0

2

4

6

8

10

12

14

16

18

Cha

nge

in s

oil N

H 4

+ (µg

N g

soi

l -1)

.

A)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Cha

nge

in s

oil N

O 3

- (µg

N g

soi

l -1)

B)

*

0

2

4

6

8

10

12

14

16

18

Cha

nge

in T

IN (µ

g N

g s

oil -1

) .

C)

108

Figure 7

0.0

0.1

0.2

0.3

0.4

0.5

Rat

e of

NH

4+ acc

umul

atio

n .

(ugN

g re

sin

-1 d

ay -1

)

A)

Salvage period

Pre- Salvage

Post- salvage

0.0

0.1

0.2

0.3

0.4

0.5

Rat

e of

NO

3- a

ccum

ualti

on .

(ug

N g

resi

n -1

day

-1)


B)

0

10

20

30

40

50

60

70

Mas

s of

NH

4+ acc

umul

ated

.

(ug

N g

resi

n -1

)

C)

0

10

20

30

40

50

60

70

Summer2007

Winter2007-08

Summer2008

Winter2008-09

Summer2009

Winter2009-10

Summer2010

Mas

s of

NO

3- a

ccum

ulat

ed .

(ug

N g

resi

n -1

)


D)

109

Figure 8

-30

-20

-10

0

10

20

30

40

50

60

Cha

nge

in a

nnua

l mas

s of

resi

n .

bag

N (u

g N

g re

sin -1

)

Beetle only Beetle + Salvage

NH4+ NO3

- TIN

110

THESIS CONCLUSIONS

Disturbances can change and shape ecosystems by creating or reducing spatial heterogeneity,

redistributing biomass among ecosystem pools, altering resource availability and abiotic

conditions, or initiating community composition change (Bormann and Likens 1979, Pickett and

White 1985, Turner and Dale 1998). In Greater Yellowstone and across western North America,

bark beetles have thus shaped conifer forests as a native component of the disturbance regime

throughout the Holocene (Brunelle et al. 2008). However, recent outbreaks have reached an

unprecedented spatial extent and severity (Raffa et al. 2008), the frequency and severity of

outbreaks are expected to increase in a warming climate (Logan et al. 2010), and there is

growing concern over the impact of beetle outbreak on ecosystem services (Flint 2006) and

interactions with other disturbance types (Simard et al. 2011). In this dissertation I have sought

to fill large knowledge gaps regarding the impact of beetle disturbance on N cycling, the role of

forest type in determining ecosystem response, and the effect of forest management on post-

beetle ecosystem function. Following are the main conclusions from this body of work:

(1) Post-beetle patterns of N cycling change in the litter, soil, and vegetation of conifer

forests were consistent with the guild-specific impacts of bark beetles on canopy

vegetation. Two to four years after disturbance, both biotic and abiotic controls were

important factors in the increase of N content in litter and soils, with concurrent increases

in the foliar N of undisturbed vegetation. By thirty years after outbreak, soil N

availability returned to undisturbed levels though the canopy N pool remained depleted.

Overall impact on soil N cycling was less severe than seen in post-fire forests.

111

(2) Contrasting conifer forest types have qualitatively similarly responses to beetle

disturbance in the litter-soil profile, though the response of undisturbed vegetation

varies. Despite substantial differences in the pre-disturbance N dynamics of lodgepole

pine and Douglas-fir forests, both showed 25-50% increases in needle litter N content and

soil N pools and fluxes. However, only in lodgepole pine did increased soil N

availability lead to increased foliar N of unattacked canopy trees.

(3) Beetle-killed forests exhibit several mechanisms of N retention following

disturbance, partially mitigating the potential for N loss via soil NO3- leaching.

Minimal increases in nitrification rate and soil NO3- pools, increased N concentration in

undisturbed canopy biomass, and declines in foliar Mn suggest multiple pathways for N

retention in beetle-killed forests.

(4) Salvage logging in beetle-killed forests has few additional impacts on N cycling in

the first year after harvest. Salvage harvest did not increase litter inputs or soil N

cycling any more than would have occurred in un-salvaged beetle-killed forests.

However, substantial differences in coarse wood inputs and understory vegetation

composition are likely to develop over longer time scales.

(5) Salvage reduced the total density of saplings in beetle-killed forests, though residual

densities should be sufficient for stand replacement. Salvage may promote the

dominance of lodgepole pine in post-beetle forests, as lodgepole pine saplings were less

susceptible to density decline following salvage than other species.

112

REFERENCES

Bormann, F. H., and G. E. Likens. 1979. Pattern and process in a forested ecosystem: disturbance, development, and the steady state based on the Hubbard Brook ecosystem study. Springer-Verlag, New York. Brunelle, A., G. E. Rehfeldt, B. Bentz, and A. S. Munson. 2008. Holocene records of Dendroctonus bark beetles in high elevation pine forests of Idaho and Montana, USA. Forest Ecology and Management 255:836-846. Flint, C. G. 2006. Community perspectives on spruce beetle impacts on the Kenai Peninsula, Alaska. Forest Ecology and Management 227:207-218. Logan, J. A., W. W. Macfarlane, and L. Willcox. 2010. Whitebark pine vulnerability to climate- driven mountain pine beetle disturbance in the Greater Yellowstone Ecosystem. Ecological Applications 20:895-902. Pickett, S. T. A., and P. S. White, editors. 1985. The ecology of natural disturbance and patch dynamics. Academic Press, Orlando, FL. Raffa, K. F., B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H. Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic amplification: The dynamics of bark beetle eruptions. Bioscience 58:501-517. Simard, M., W. H. Romme, J. M. Griffin, and M. G. Turner. 2011. Do mountain pine beetle outbreaks change the probability of active crown fire in lodgepole pine forests? Ecological Monographs 81:3-24. Turner, M. G., and V. H. Dale. 1998. Comparing large, infrequent disturbances: What have we learned? Ecosystems 1:493-496.

bark beetle disturbance and nitrogen cycling in … · 2012-05-01 · attack (safranyik and carroll...

Documents