Sotto, M. Jonas (2024) LEVERAGING CONVERGENT CROSS MAPPING TO UNCOVER LONG-DISTANCE CAUSAL INFLUENCES PRE - Research Project, ENSTA.
![]()
| PDF 2088Kb |
Abstract
I investigate the application of Convergent Cross Mapping (CCM) to explore potential cas cading effects of regime shifts (RS) across distant ecosystems, focusing on the Amazon region. My research builds on the work of the Stockholm Resilience Centre and specifically targets the detection of causal relationships in complex systems through time series analysis. Utilizing global datasets on Gross Primary Productivity (GPP) as a proxy for resilience loss (RL), I examine the influence of RS from one region on another, hypothesizing that changes in the Amazon rainforest could impact ecosystems to its west through moisture recycling. Results indicated varying out comes depending on the dataset used, with raw monthly data suggesting potential causal links, whereas deseasonalized data yielded inconclusive results, possibly due to the impact of seasonality on the CCM algorithm. I conclude that while CCM offers a promising approach to understanding global teleconnections, further research is necessary, particularly in refining the methodology, ex ploring different datasets, and improving the understanding of RL influences on RS across spatial scales.
Item Type: | Thesis (PRE - Research Project) |
---|---|
Uncontrolled Keywords: | environnement, analyse de séries temporelles, écologie, basculement de régime, système socio-écologique, causalité, algorithme |
Subjects: | Information and Communication Sciences and Technologies Earth Sciences and Environmental Engineering Life Sciences and Engineering |
ID Code: | 10156 |
Deposited By: | Jonas SOTTO |
Deposited On: | 27 août 2024 18:24 |
Dernière modification: | 16 sept. 2024 14:34 |
Repository Staff Only: item control page