Wind damage propagation in forests

Damage propagation, Forest wind damage, Tree motion, Wind risk

Download the article:

https://drive.google.com/file/d/0ByVlxqWw8eVkZmtFd1BNWUJnQWM/view?usp=sharing

Contributions : 

1-Dupont, Brunet designed the research

2-Pivato deeply modified the research, Dupont was against these changes
3-Pivato worked during more than 5 years including 18 months without funding

4-Pivato carried out the first model simulations and data analysis

5-Pivato wrote his PhD dissertation

6-Dupont, Brunet contributed ideas to the dissertation

7-Pivato defended his PhD and quit the INRA (May 2014)

8-Pivato worked for the publication of the 3^rd chapter of his dissertation

9-Dupont, Brunet took Pivato apart from the publication of his work
10-Dupont improved the model simulations and data analysis
11-Brunet contributed ideas to the data analysis
12-Dupont, Brunet drafted the manuscript and of course Dupont made himself first author

13-Dupont, Brunet submitted the manuscript to Nature (January 2015) without the agreement of Pivato

14-Nature refused the manuscript without reviewing it

15-Dupont, Brunet submitted the manuscript to Nature communication

16-Nature communication refused the manuscript without reviewing it

17-Dupont, Brunet submitted the manuscript to PNAS

18-PNAS refused the manuscript without reviewing it

19-Dupont, Brunet submitted the manuscript to AFM

20-The manuscript was reviewed and sent back for revision

21-Pivato dissented from the way the manuscript was drafted

22-Dupont, Brunet called upon the INRA legal department to defend them because they were blindly persuaded to be right. They lied by saying Pivato did not contribute to the writing of the manuscript.
23- The INRA legal department forced Dupont, Brunet to add the mention “Dupont, Pivato contributed equally to this work”

24- The manuscript was sent back to the AFM and accepted

The purpose of this webpage is to warn people.

As I saw pain and tears in the eyes of a good person who was despoiled by Yves Brunet.

As my work was cowardly recovered and exploited by Sylvain Dupont and Yves Brunet.

Be careful if you plan to work with these two usurpers.

The Code

Here is the f90 code of the model described in Pivato et al, 2014. This version provides the animation described at the end of the article ( .plt file, use tecplot to post process)

https://drive.google.com/file/d/0ByVlxqWw8eVkaEYxNHlTbHlDMUk/view?usp=sharing

Phases of damage propagation during a windstorm

Here is a slide from my PhD defence, showing the different evolution phases of damage propagation during a windstorm ( Same simulation as the animations published here the 19/11/2014 )

And here is a serie of results (from chapter 4 of my PhD dissertation) showing the two main phases ( 1 - damage start/ first minor gust (dashed line) and 2 - damage propagation acceleration/ more important gust (continuous line). Phases 1 and 2 can be the same, phase 2 can never occured.

These different cases are related to a study concerning the effect of foliar area reduction on damage propagation.

Modelling wind damage propagation in a homogeneous forest

This is chapter 3 of my PhD dissertation (also available on ResearchGate):

https://drive.google.com/file/d/0ByVlxqWw8eVkRTgyRXFFNVVHYzQ/view?usp=sharing

Abstract: In order to investigate wind-tree interaction and tree damage propagation in windstorm conditions, a tree swaying model accounting for large tree deflections and considering the possibility of tree breakage is coupled with a large-eddy simulation airflow model. The model is applied on a homogeneous Maritime pine forest under windstorm conditions. It is first shown that the well-known characteristics of canopy flow remain identical in windstorm conditions and that tree motion has a negligible impact on the wind dynamics. Then, tree damage propagation during windstorms is analysed in relation to windstorm intensity. Under similar conditions the level of damages can differ significantly between two identical stands, depending on the occurrence of critical gusts. Once the first damages are initiated, the velocity of damage propagation is similar for windstorms of similar intensity and increases exponentially with windstorm intensity. This is the first time that tree damages are explicitly simulated using an airflow model applied in windstorm conditions.


Keywords: Fluid-structure interaction - Tree motion - Turbulent flow - Large eddy simulation - Wind damage - Windstorm.

Tree motion model animations

Here are 2 animations of my tree motion model. Each stem is a 21m heigh maritime pine. The wind is made with sinusoidal functions.

Case of 2 forests in a fragmented landscape

Case of 4 forests in a fragmented landscape (see end of chapter 2 in my phd dissertation for more details about this video)

Quelques animations...

Création d'une trouée (en rouge) puis propagation des dégâts et modification de l'écoulement du vent.
Creation of a gap (red area) then damage propagation and modification of the wind flow.

Ma thèse - My PhD thesis

Les interactions vent-forêt en condition de tempête : 

un modèle couplé prenant en compte la rupture des arbres

- Résumé : La connaissance de la réponse au vent des arbres est importante pour 

mieux comprendre l'impact d'une tempête sur un massif forestier et identifier 

des stratégies visant à limiter les dégâts. Pour améliorer notre compréhension 

de l'interaction vent-plantes à l'échelle du paysage, la modélisation mathématique 

et mécanique de cette interaction est indispensable, la multiplication de mesures 

ponctuelles dans un environnement hétérogène étant difficilement réalisable. 

Ainsi, cette thèse présente d'abord le développement d'un modèle original et 

novateur du mouvement de l'arbre permettant de simuler les grandes déflexions 

se produisant lors de tempêtes, et ce jusqu'à la rupture, le tout avec 

suffisamment de simplicité dans la simulation du mouvement de forêts 

contenant un grand nombre d'arbres. Ensuite, ce modèle a été couplé à un 

modèle d'écoulement atmosphérique pour représenter les conditions de tempêtes 

et a permis de simuler l'interaction vent-forêt à l'échelle du paysage. 

Pour la première fois, la création dynamique et l'expansion de trouées dans 

une forêt, avec la modification de l'écoulement du vent qui en résulte, ont 

été simulées de manière réaliste.

La suite de ce travail a montré le potentiel de notre modèle pour étudier 

l'impact des pratiques de gestion forestière sur la vulnérabilité des forêts 

face aux tempêtes.

En conclusion, ce nouveau modèle apporte des avancées significatives dans la 

modélisation du mouvement de forêt et dans la compréhension des phénomènes 

en jeu lors d'une tempête. Cette connaissance devrait être une bonne base pour 

la définition de meilleures pratiques de gestion des forêts.

Mots clés : Dégâts, Interaction fluide-structure, Mouvement de l'arbre, 

Simulation des grandes échelles, Tempête, Turbulence, Vent.

Ma thèse : 

https://drive.google.com/file/d/0ByVlxqWw8eVkM3R5VDdHUTEzZFE/view?usp=sharing 

_____________

Wind-forest interactions in windstorm conditions : 

a coupled model with account for tree breakage

- Abstract : To know how trees behave under windload is important to understand 

the impact of a windstorm on a forest and identify strategies for limiting wind 

damage. In order to improve our understanding of wind-tree interaction at the 

landscape scale, mathematical and mechanical modelling of this interaction is 

essential, as performing numerous specific measurements in a heterogeneous 

environment is not feasible.

Thus, this thesis first presents the development of an original and innovative 

tree motion model able to simulate large deflections occurring during storms. 

This model allows trees to break and has sufficient simplicity to allow 

simulations of forest motion containing a large number of trees. 

This model was then coupled to an atmospheric flow model in order to represent 

wind storm conditions. It has been used to simulate wind-forest interaction 

at landscape scale. For the first time, the dynamic creation and propagation 

of gaps in a forest have been simulated realistically, along with the resulting 

wind flow modifications. Finally, further studies have demonstrated the 

efficiency of our model for studying the impact of forest management practices 

on forest vulnerability to storms.

In conclusion, this new model has allowed significant advances in forest motion 

modelling and in the understanding of the phenomena involved during a windstorm. 

This knowledge should be a good basis for defining best practices in forestry 

management.

Keywords : Fluid-structure interaction, Large eddy simulation, Tree motion, Turbulent flow, Wind, Wind damage, Windstorm.

My PhD thesis (chapter 2 and 3 are in English) :
https://drive.google.com/file/d/0ByVlxqWw8eVkM3R5VDdHUTEzZFE/view?usp=sharing