Abstract
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.
Highlights
Approaches which aim to define regions with similar biophysical characteristics are commonly known as land cover classification schemes and are widely used in multiple geoscientific disciplines
For the alternative structure optimization (ASO)-MTL method, we have experimented with the value of the h parameter, which is the dimensionality of the shared feature space – see Sect. 3.2 for more details about the influence of this parameter on the clustering results
4 Conclusion In this paper we introduced a novel framework for identifying regions with similar biosphere–climate interplay dynamics
Summary
Approaches which aim to define regions with similar biophysical characteristics are commonly known as land cover classification schemes and are widely used in multiple geoscientific disciplines. Traditional land use and/or land cover (change) classifications are typically based on spectral information from the land surface coming from satellites (Loveland and Belward, 1997; Congalton et al, 2014). Amongst the most well known and widely used are the International Geosphere– Biosphere Program DISCover Global 1 km Land Cover classification (IGBP-DIS) (Loveland et al, 2000), Global Land Cover 2000 (Bartholomé and Belward, 2005), and more recently the land cover map developed within the European Space Agency’s Climate Change Initiative (ESA CCI) (Poulter et al, 2015; Li et al, 2018). Dynamics in these climate regimes are used as a diagnostic of climate change by exploring their shifting boundaries (e.g. Diaz and Eischeid, 2007; Chen and Chen, 2013; Zhang and Yan, 2014a, b; Spinoni et al, 2015; Chan and Wu, 2015) or as a means to predict future climatic zone distributions using climate projections (e.g. Hanf et al, 2012, Gallardo et al, 2013, Mahlstein et al, 2013)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
