Advanced search
1 file | 2.27 MB Add to list

Global hydro-climatic biomes identified via multitask learning

(2018) GEOSCIENTIFIC MODEL DEVELOPMENT. 11(10). p.4139-4153
Author
Organization
Abstract
The most widely used global land cover and climate classifications are based on vegetation characteristics and/or climatic conditions derived from observational data. However, these classification schemes do not directly stem from the characteristic interaction between the local climate and the biotic environment. In this work, we model the dynamic interplay between vegetation and local climate in order to delineate ecoregions that share a coherent response to hydro-climate variability. Our novel framework is based on a multitask learning approach that discovers the spatial relationships among different locations by learning a low-dimensional representation of predictive structures. This low-dimensional representation is combined with a clustering algorithm that yields a classification of biomes with coherent behaviour. Experimental results using global observation-based datasets indicate that, without the need to prescribe any land cover information, the identified regions of coherent climate-vegetation interactions agree well with the expectations derived from traditional global land cover maps. The resulting global "hydro-climatic biomes" can be used to analyse the anomalous behaviour of specific ecosystems in response to climate extremes and to benchmark climate-vegetation interactions in Earth system models.
Keywords
GEOGRAPHICALLY WEIGHTED REGRESSION, WORLD MAP, CLASSIFICATION, VEGETATION, FRAMEWORK, INVESTIGATE, UNCERTAINTY, INCREASES, CONTEXT, MODELS

Downloads

  • KERMIT-A1-480.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 2.27 MB

Citation

Please use this url to cite or link to this publication:

MLA
Papagiannopoulou, Christina, et al. “Global Hydro-Climatic Biomes Identified via Multitask Learning.” GEOSCIENTIFIC MODEL DEVELOPMENT, vol. 11, no. 10, 2018, pp. 4139–53, doi:10.5194/gmd-11-4139-2018.
APA
Papagiannopoulou, C., Miralles, D., Demuzere, M., Verhoest, N., & Waegeman, W. (2018). Global hydro-climatic biomes identified via multitask learning. GEOSCIENTIFIC MODEL DEVELOPMENT, 11(10), 4139–4153. https://doi.org/10.5194/gmd-11-4139-2018
Chicago author-date
Papagiannopoulou, Christina, Diego Miralles, Matthias Demuzere, Niko Verhoest, and Willem Waegeman. 2018. “Global Hydro-Climatic Biomes Identified via Multitask Learning.” GEOSCIENTIFIC MODEL DEVELOPMENT 11 (10): 4139–53. https://doi.org/10.5194/gmd-11-4139-2018.
Chicago author-date (all authors)
Papagiannopoulou, Christina, Diego Miralles, Matthias Demuzere, Niko Verhoest, and Willem Waegeman. 2018. “Global Hydro-Climatic Biomes Identified via Multitask Learning.” GEOSCIENTIFIC MODEL DEVELOPMENT 11 (10): 4139–4153. doi:10.5194/gmd-11-4139-2018.
Vancouver
1.
Papagiannopoulou C, Miralles D, Demuzere M, Verhoest N, Waegeman W. Global hydro-climatic biomes identified via multitask learning. GEOSCIENTIFIC MODEL DEVELOPMENT. 2018;11(10):4139–53.
IEEE
[1]
C. Papagiannopoulou, D. Miralles, M. Demuzere, N. Verhoest, and W. Waegeman, “Global hydro-climatic biomes identified via multitask learning,” GEOSCIENTIFIC MODEL DEVELOPMENT, vol. 11, no. 10, pp. 4139–4153, 2018.
@article{8578971,
  abstract     = {{The most widely used global land cover and climate classifications are based on vegetation characteristics and/or climatic conditions derived from observational data. However, these classification schemes do not directly stem from the characteristic interaction between the local climate and the biotic environment. In this work, we model the dynamic interplay between vegetation and local climate in order to delineate ecoregions that share a coherent response to hydro-climate variability. Our novel framework is based on a multitask learning approach that discovers the spatial relationships among different locations by learning a low-dimensional representation of predictive structures. This low-dimensional representation is combined with a clustering algorithm that yields a classification of biomes with coherent behaviour. Experimental results using global observation-based datasets indicate that, without the need to prescribe any land cover information, the identified regions of coherent climate-vegetation interactions agree well with the expectations derived from traditional global land cover maps. The resulting global "hydro-climatic biomes" can be used to analyse the anomalous behaviour of specific ecosystems in response to climate extremes and to benchmark climate-vegetation interactions in Earth system models.}},
  author       = {{Papagiannopoulou, Christina and Miralles, Diego and Demuzere, Matthias and Verhoest, Niko and Waegeman, Willem}},
  issn         = {{1991-959X}},
  journal      = {{GEOSCIENTIFIC MODEL DEVELOPMENT}},
  keywords     = {{GEOGRAPHICALLY WEIGHTED REGRESSION,WORLD MAP,CLASSIFICATION,VEGETATION,FRAMEWORK,INVESTIGATE,UNCERTAINTY,INCREASES,CONTEXT,MODELS}},
  language     = {{eng}},
  number       = {{10}},
  pages        = {{4139--4153}},
  title        = {{Global hydro-climatic biomes identified via multitask learning}},
  url          = {{http://dx.doi.org/10.5194/gmd-11-4139-2018}},
  volume       = {{11}},
  year         = {{2018}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: