Disturbance Effect Estimation Methods (DEEM)

Sampling Strategies for Forest Disturbance Estimation

Overview

This project is part of the focal project X-RISKS of the French national research programme PEPR FORESTT. The aim is to develop sampling strategies for estimating the effects of multiple disturbances in forests, such as volume or biomass loss due to storms or wildfires, and tree mortality or decline in growth due to insect or pathogen infestations or droughts.

Objectives

• Develop sampling strategies (sampling design and estimators) for the French National Forest Inventory (NFI).
• Utilize auxiliary data correlated with disturbance manifestations, such as remote sensing data or spatially interpolated meteorological data.
• Create sampling designs using methods of spatially balanced sampling and continuous-population importance sampling.
• Investigate design-based model-assisted estimators for improving the precision of disturbance effect estimations.
• Test strategies via simulations in a realistic, large-scale artificial tree population ("digital twin" of a 3600-km² region in North-Eastern France).

Methodology

• Sampling Design and Estimation: Proposing methods for creating a sampling design and model-assisted estimator to estimate the effect of individual disturbances.
• Simulation Testing: Participating in the creation of tree attributes and artificial auxiliary data in the simulation population, and testing proposed sampling strategies.
• Multiple Disturbances: Reflecting on addressing multiple simultaneous disturbances overlapping in space.
• Integration with NFI: Reflecting on incorporating information obtained from disturbance effect estimation into the standard estimation of the French NFI.

Collaboration

The project is supervised by Minna Pulkkinen (Laboratory of Forest Inventory LIF, Nancy, France) and involves collaboration with:

• Alexander Massey (Laboratory of Forest Inventory LIF, Nancy, France)
• Guillaume Chauvet (ENSAI, Rennes, France)
• Olivier Bouriaud (University of Suceava, Romania)
• Cédric Vega (Laboratory of Forest Inventory LIF, Nancy, France)

Tasks

1. Conduct a literature review on methods utilizing auxiliary data in sampling design and estimation.
2. Propose auxiliary-data-based methods for creating a sampling design and model-assisted estimator.
3. Participate in creating tree attributes and artificial auxiliary data for simulation testing.
4. Reflect on addressing multiple simultaneous disturbances and integrating disturbance effect estimation with the French NFI.

Outputs

• Scientific articles on the results of the project.
• Sampling strategies compatible with the French NFI framework.
• Demonstrate with working examples in R.

Getting Started

To get started with this project, clone the repository and explore the R scripts and data files provided.

git clone [https://github.com/afmass/DEEM]

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

For more information, please contact Alexander Massey at [alexander.massey@ign.fr].