Modelling forest ruin due to climate hazards

Modelling forest ruin due to climate hazards

Pascal Yiou (1, 2) , Nicolas Viovy (1, 3)

1 LSCE - Laboratoire des Sciences du Climat et de l'Environnement [Gif-sur-Yvette]
2 ESTIMR - Extrèmes : Statistiques, Impacts et Régionalisation
3 MOSAIC - Modélisation des Surfaces et Interfaces Continentales

Estimating the risk of forest collapse due to extreme climate events is one of the challenges of adapting to climate change. We adapt a concept from ruin theory, which is widely used in econometrics and the insurance industry, to design a growth-ruin model for trees which accounts for climate hazards that can jeopardize tree growth. This model is an elaboration of a classical Cramer-Lundberg ruin model that is used in the insurance industry. The model accounts for the interactions between physiological parameters of trees and the occurrence of climate hazards. The physiological parameters describe interannual growth rates and how trees react to hazards. The hazard parameters describe the probability distributions of the occurrence and intensity of climate events. We focus on a drought-heatwave hazard. The goal of the paper is to determine the dependence of the forest ruin and average growth probability distributions on physiological and hazard parameters. Using extensive Monte Carlo experiments, we show the existence of a threshold in the frequency of hazards beyond which forest ruin becomes certain to occur within a centennial horizon. We also detect a small effect of the strategies used to cope with hazards. This paper is a proof of concept for the quantification of forest collapse under climate change.

Domaines: Océan, Atmosphère; Interfaces continentales, environnement

Pascal Yiou, Nicolas Viovy. Modelling forest ruin due to climate hazards. Earth System Dynamics, 2021, 12 (3), pp.997 - 1013. ⟨10.5194/esd-12-997-2021⟩. ⟨hal-03351183⟩