Browsing by Subject "boosted regression trees"

Sort by: Order: Results:

Now showing items 1-4 of 4
  • Pajunen, Virpi; Jyrkänkallio-Mikkola, Jenny; Luoto, Miska; Soininen, Janne (2019)
    Species occurrences are influenced by numerous factors whose effects may be context dependent. Thus, the magnitude of the effects and their relative importance to species distributions may vary among ecosystems due to anthropogenic stressors. To investigate context dependency in factors governing microbial bioindicators, we developed species distribution models (SDMs) for epilithic stream diatom species in human-impacted and pristine sites separately. We performed SDMs using boosted regression trees for 110 stream diatom species, which were common to both data sets, in 164 human-impacted and 164 pristine sites in Finland (covering similar to 1,000 km, 60 degrees to 68 degrees N). For each species and site group, two sets of models were conducted: climate model, comprising three climatic variables, and full model, comprising the climatic and six local environmental variables. No significant difference in model performance was found between the site groups. However, climatic variables had greater importance compared with local environmental variables in pristine sites, whereas local environmental variables had greater importance in human-impacted sites as hypothesized. Water balance and conductivity were the key variables in human-impacted sites. The relative importance of climatic and local environmental variables varied among individual species, but also between the site groups. We found a clear context dependency among the variables influencing stream diatom distributions as the most important factors varied both among species and between the site groups. In human-impacted streams, species distributions were mainly governed by water chemistry, whereas in pristine streams by climate. We suggest that climatic models may be suitable in pristine ecosystems, whereas the full models comprising both climatic and local environmental variables should be used in human-impacted ecosystems.
  • Virtanen, Elina A.; Viitasalo, Markku; Lappatainen, Juho; Moilanen, Atte (2018)
  • Fang, Keyan; Frank, David; Zhao, Yan; Zhou, Feifei; Seppä, Heikki (2015)
    Investigations of climate-growth interactions can shed light on the response of forest growth to climate change and the dendroclimatic reconstructions. However, most existing studies in the climatically important Tibetan Plateau (TP) and surrouding regions focus on linear growth responses to environmental variation. Herein we investigated both the linear and the nonlinear climate-growth interactions for 152 tree-ring chronologies in the TP and vicinity. Weintroduced the boosted regression tree (BRT) technique to study the nonlinear climate-growth relationships by pooling several sites with similar climate-growth relationships to mitigate potential biases due to the shortness of the instrumental records. Across most of the TP and surroundings, tree growth is stressed by drought. The warming induced drought has been evidenced by the strong interactions between temperature and precipitation in the BRT analyses. The drought stress on forest growth is particularly conspicuous for a hydrological year over much of the Northern TP and surroundings. The BRT analyses indicate the compensation effect of moisture prior to the growing season for the moisture deficit in the early growing season in May to July, when most of the ring-width formation occurs.