학술논문
A non-linear data-driven approach to reveal global vegetation sensitivity to climate
Document Type
Conference
Author
Source
2017 9th International Workshop on the Analysis of Multitemporal Remote Sensing Images (MultiTemp) Analysis of Multitemporal Remote Sensing Images (MultiTemp), 2017 9th International Workshop on the. :1-3 Jun, 2017
Subject
Language
Abstract
Following a Granger causality framework based on a random forest predictive model, we exploit the current wealth of multi-decadal satellite data records to uncover the main spatiotemporal drivers of monthly vegetation variability globally. Results based on 1981–2010 indicate that water availability is the most dominant factor driving vegetation globally. This overall dependency of vegetation on water availability is larger than previously reported, partly owed to the ability of the framework to disentangle the co-linearites between climate drivers and to quantify non-linear impacts of climate on vegetation. This is a first step towards a quantitative comparison of the resistance and resilience of different ecosystems, and can be used to benchmark climate model representation of vegetation sensitivity.