The  Changing  Structure  and  Function  of  Arthropod  Food  Webs   in  a  Warming  Arctic   by   Amanda  May  Koltz   University  Program  in  Ecology   Duke  University     Date:_______________________   Approved:     ___________________________   Justin  Wright,  Supervisor     ___________________________   William  Morris,  Chair     ___________________________   Dean  Urban     ___________________________   Robert  Dunn       Dissertation  submitted  in  partial  fulfillment  of   the  requirements  for  the  degree  of  Doctor  of  Philosophy     in  the  Program  in  Ecology  in  the  Graduate  School  of   Duke  University     2015       i v ABSTRACT   The  Changing  Structure  and  Function  of  Arthropod  Food  Webs   in  a  Warming  Arctic   by   Amanda  May  Koltz   University  Program  in  Ecology   Duke  University     Date:_______________________   Approved:     ___________________________   Justin  Wright,  Supervisor     ___________________________   William  Morris,  Chair     ___________________________   Dean  Urban     ___________________________   Robert  Dunn       An  abstract  of  a  dissertation  submitted  in  partial   fulfillment  of  the  requirements  for  the  degree   of  Doctor  of  Philosophy  in  the  Program  in  Ecology     in  the  Graduate  School  of   Duke  University     2015   i v Copyright  by   Amanda  May  Koltz   2015 Abstract Environmental  changes,  such  as  climate  change,  can  have  differential  effects  on   species,  with  important  consequences  for  community  structure  and  ultimately,  for   ecosystem  functioning.  In  the  Arctic,  where  ecosystems  are  experiencing  warming  at   twice  the  rate  as  elsewhere,  these  effects  are  expected  to  be  particularly  strong.  A  proper   characterization  of  the  link  between  warming  and  biotic  interactions  in  these  particular   communities  is  of  global  importance  because  the  tundra’s  permafrost  stores  a  vast   amount  of  carbon  that  could  be  released  through  decomposition  as  greenhouse  gases   and  alter  the  global  rate  of  climate  change.  In  this  dissertation,  I  examine  how  arthropod   communities  are  responding  to  warming  in  the  Arctic  and  how  these  responses  might  be   affecting  ecosystem  functioning.     I  first  address  the  question  of  whether  and  how  long-­‐‑term  changes  in  climate  are   affecting  individual  groups  and  overall  community  structure  in  a  high-­‐‑arctic  arthropod   food  web.  I  find  that  increasingly  warm  springs  and  summers  between  1996-­‐‑2011   differentially  affected  some  arthropod  groups  and  that  this  led  to  major  changes  in  the   relative  abundances  of  different  trophic  groups  within  the  arthropod  community.   Specifically,  spring  and  summer  warming  are  associated  with  higher  absolute   abundances  of  herbivores  and  parasitoids  and  lower  relative  abundances  of  detritivores   within  the  community.  These  changes  are  particularly  pronounced  in  heath  sites,     iv suggesting  that  arthropod  communities  in  dry  habitats  are  more  responsive  to  climate   change  than  those  in  wet  habitats.  I  also  show  that  herbivores  and  parasitoids  are   sensitive  to  conditions  at  subzero  temperatures,  even  during  periods  of  diapause,  and   that  all  trophic  groups  benefit  from  a  longer  transition  period  between  summer  and   winter.  These  results  suggest  that  the  projected  winter  and  springtime  warming  in   Greenland  may  have  unexpected  consequences  for  northern  arthropod  communities.   Moreover,  the  relative  increase  in  herbivores  and  loss  of  detritivores  may  be  changing   the  influence  of  the  arthropod  community  over  key  ecosystem  processes  such  as   decomposition,  nutrient  cycling,  and  primary  productivity  in  the  tundra.     Predator-­‐‑induced  trophic  cascades  have  been  shown  to  impact  both  community   structure  and  ecosystem  processes,  yet  it  is  unclear  how  climate  change  may  exacerbate   or  dampen  predator  effects  on  ecosystems.  In  the  second  chapter  of  my  dissertation,  I   investigate  the  role  of  one  of  the  dominant  tundra  predators  within  the  arctic  ecosystem,   wolf  spiders,  and  how  their  impact  might  be  changing  with  warming.  Using  results   from  a  two-­‐‑year-­‐‑long  field  experiment,  I  test  the  influence  of  wolf  spider  density  over   the  structure  of  soil  microarthropod  communities  and  decomposition  rates  under  both   ambient  and  artificially  warmed  temperatures.  I  find  that  predator  effects  on  soil   microarthropods  change  in  response  to  warming  and  that  these  changes  translate  into   context-­‐‑specific  indirect  effects  of  predators  on  decomposition.  Specifically,  while  high   densities  of  wolf  spiders  lead  to  faster  decomposition  rates  at  ambient  temperatures,     v they  are  associated  with  slower  decomposition  rates  in  experimentally  warmed  plots.   My  results  suggest  that  if  warming  causes  an  increase  in  arctic  wolf  spider  densities,   these  spiders  may  buffer  the  rate  at  which  the  massive  pool  of  stored  carbon  is  lost  from   the  tundra.     Wolf  spiders  in  the  Arctic  are  expected  to  become  larger  with  warming,  but  it  is   unclear  how  this  change  in  body  size  will  affect  spider  populations  or  the  role  of  wolf   spiders  within  arctic  food  webs.  In  the  third  chapter  of  my  dissertation,  I  explore  wolf   spider  population  structure  and  juvenile  recruitment  at  three  sites  of  the  Alaskan  Arctic   that  naturally  differ  in  mean  spider  body  size.  I  find  that  there  are  fewer  juveniles  in   sites  where  female  body  sizes  are  larger  and  that  this  pattern  is  likely  driven  by  a  size-­‐‑ related  increase  in  the  rate  of  intraspecific  cannibalism.  These  findings  suggest  that   across  the  tundra  landscape,  there  is  substantial  variation  in  the  population  structure   and  trophic  position  of  wolf  spiders,  which  is  driven  by  differences  in  female  spider   body  sizes.     Overall,  this  dissertation  demonstrates  that  arctic  arthropod  communities  are   changing  as  a  result  of  warming.  In  the  long-­‐‑term,  warming  is  causing  a  shift  in   arthropod  community  structure  that  is  likely  altering  the  functional  role  of  these  animals   within  the  ecosystem.  However  even  in  the  short-­‐‑term,  warming  can  alter  species   interactions  and  community  structure,  with  important  consequences  for  ecosystem   function.  Arthropods  are  not  typically  considered  to  be  major  players  in  arctic     vi ecosystems,  but  I  provide  evidence  that  this  assumption  should  be  questioned.   Considering  that  they  are  the  largest  source  of  animal  biomass  across  much  of  the   tundra,  it  is  likely  that  their  activities  have  important  consequences  for  regional  and   global  carbon  dynamics.       vii Dedication This  dissertation  is  dedicated  to  my  parents,  stepparents,  and  husband.         viii Contents Abstract  ..........................................................................................................................................  iv   List  of  Tables  ................................................................................................................................  xii   List  of  Figures  .............................................................................................................................  xiv   Acknowledgements  ...................................................................................................................  xvi   1.  Introduction  ...............................................................................................................................  1   1.1  Species  and  community  responses  to  climate  change  .................................................  1   1.2  Ecological  consequences  of  climate  change  in  the  Arctic  ...........................................  1   1.3  Arthropods  as  a  model  system  for  understanding  effects  of  climate  change  on   biological  communities  ..........................................................................................................  3   1.4  Objectives  ...........................................................................................................................  5   2.  Differential  response  to  climate  change  among  arthropods  is  altering  the  structure  of   arctic  communities  ........................................................................................................................  7   2.1  Introduction  .......................................................................................................................  7   2.2  Methods  ...........................................................................................................................  11   2.2.1  Study  sites  ...................................................................................................................  11   2.2.2  Climate  data  ...............................................................................................................  11   2.2.3  Arthropod  data  collection  ........................................................................................  12   2.2.4  Analyses  ......................................................................................................................  15   2.3  Results  ..............................................................................................................................  16   2.3.1  Climate  change  over  the  study  period  ...................................................................  16   2.3.2  Links  between  arthropod  abundances  and  environmental  predictors  .............  18     ix 2.3.3  Links  between  variability  in  community  structure  and  environmental   predictors  .............................................................................................................................  18   2.4  Discussion  ........................................................................................................................  25   3.  Predator  effects  on  a  soil  food  web  can  buffer  carbon  losses  in  a  warmer  Arctic  .........  31   3.1  Introduction  .....................................................................................................................  31   3.2  Methods  ...........................................................................................................................  34   3.2.1  Experimental  design  .................................................................................................  34   3.2.2  Measures  of  community  structure  and  function  ..................................................  36   3.2.2.1  Sampling  the  surface  and  belowground  community  ...................................  37   3.2.2.2  Decomposition  ....................................................................................................  37   3.2.2.3  Soil  moisture  .......................................................................................................  38   3.2.3  Statistical  analyses  .....................................................................................................  38   3.3  Results  ..............................................................................................................................  39   3.3.1  Effects  of  wolf  spider  density  and  temperature  on  microarthropods  ...............  39   3.3.2  Effects  of  wolf  spider  density  and  temperature  on  decomposition  ...................  41   3.4  Discussion  ........................................................................................................................  44   4.  Larger  female  body  sizes  lead  to  increased  cannibalism  rather  than  higher  recruitment   in  arctic  wolf  spiders  ...................................................................................................................  50   4.1  Summary  ..........................................................................................................................  50   4.2  Results  and  discussion  ...................................................................................................  52   4.3  Methods  ...........................................................................................................................  61   4.3.1  Field  sampling  ............................................................................................................  61   4.3.2  Sample  processing  .....................................................................................................  63     x