Login / Signup

Amphibian and Reptilian Chorotypes in the Arid Land of Central Asia and Their Determinants.

Lu ZhouTao LiangLei Shi
Published in: Scientific reports (2019)
The analysis of the biogeographic distribution of species is the basis for establishing a strategy for land management and responding to climatic change, but research on the distribution of amphibians and reptiles in the arid land in the middle of Asia is extremely limited. After classifying the chorotypes of amphibians and reptiles in the arid land of Central Asia using a clustering analysis, we delineated their distribution characteristics and discovered the ecological determinants for the chorotypes in terms of feature selection and the Akaike information criterion (AIC). We identified 6 chorotypes at the higher level and 16 sub-chorotypes at the lower level. Compared to small-scale or subjective research, which produces unstable results, research characterized by both large scale and clustering methods yields more consistent and stable results. Our results show that the Mean Altitude (MA), Mean Annual Temperature (MAT), and Mean Temperature in the Wettest Quarter (MTWE) are the critical variables determining the higher-level chorotypes. Furthermore, geographical factors appear to have a stronger influence on chorotypes than climatic factors. Several climatic variables and MA were identified as the best fit in the AIC model at the lower level, while the sub-chorotypes are determined more by multiple climatic factors with complex relationships. The research on amphibian and reptilian distribution patterns will shed light on the overall distribution of other species in the same understudied area. Widespread species in the study area are not clearly distinguished due to the cluster analysis computing process. This problem however, appears in studies of the distribution of other organisms thus warrants further research. Our methodology based on the selection of multiple models is effective to explore how the environment determines the distributions of different animal groups.
Keyphrases
  • climate change
  • machine learning
  • single cell
  • multidrug resistant