Mandallaz Sampling Techniques for Forest Inventories

Erscheinungsjahr 2007
ISBN: 978-1-58488-977-9
Verlag: Taylor & Francis
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 272 Seiten

Reihe: Chapman & Hall/CRC Applied Environmental Statistics

Sound forest management planning requires cost-efficient approaches to optimally utilize given resources. Emphasizing the mathematical and statistical features of forest sampling to assess classical dendrometrical quantities, Sampling Techniques for Forest Inventories presents the statistical concepts and tools needed to conduct a modern forest inventory. The book first examines design-based survey sampling and inference for finite populations, covering inclusion probabilities and the Horvitz–Thompson estimator, followed by more advanced topics, including three-stage element sampling and the model-assisted estimation procedure. The author then develops the infinite population model/Monte Carlo approach for both simple and complex sampling schemes. He also uses a case study to reveal a variety of estimation procedures, relies on anticipated variance to tackle optimal design for forest inventories, and validates the resulting optimal schemes with data from the Swiss National Forest Inventory. The last chapters outline facts pertaining to the estimation of growth and introduce transect sampling based on the stereological approach. Containing many recent developments available for the first time in book form, this concise and up-to-date work provides the necessary theoretical and practical foundation to analyze and design forest inventories.

Zielgruppe

Biostatisticians, statisticians, ecologists, environmental scientists, governmental agencies and NGOs involved in forest inventories, and students of forest science.

Autoren/Hrsg.

Mandallaz, Daniel

Fachgebiete

• Naturwissenschaften Agrarwissenschaften Agrarwissenschaften

Inhaltsverzeichnis

preface
Introduction and terminology
Sampling finite populations: the essentials
Sampling schemes and inclusion probabilities
The Horvitz–Thompson estimator
Simple random sampling without replacement
Poisson sampling
Unequal probability sampling with replacement
Estimation of ratios
Stratification and post-stratification
Two-stage sampling
Single-stage cluster-sampling
Systematic sampling
Exercises
Sampling finite populations: advanced topics
Three-stage element sampling
Abstract nonsense and elephants
Model-assisted estimation procedures
Exercises
Forest Inventory: one-phase sampling schemes
Generalities
One-phase one-stage simple random sampling scheme
One-phase one-stage cluster random sampling scheme
One-phase two-stage simple random sampling
One-phase two-stage cluster random sampling
Exercises
Forest Inventory: two-phase sampling schemes
Two-phase one-stage simple random sampling
Two-phase two-stage simple random sampling
Two-phase one-stage cluster random sampling
Two-phase two-stage cluster random sampling
Internal linear models in two-phase sampling
Remarks on systematic sampling
Exercises
Forest Inventory: advanced topics
The model-dependent approach
Model-assisted approach
Small-area estimation
Modeling relationships
Exercises
Geostatistics
Variograms
Ordinary Kriging
Kriging with sampling error
Double Kriging for two-phase sampling schemes
Exercises
CASE STUDY
Optimal sampling schemes for forest inventory
Preliminaries
Anticipated variance under the local Poisson model
Optimal one-phase one-stage sampling schemes
Discrete approximations of PPS
Optimal one-phase two-stage sampling schemes
Optimal two-phase sampling schemes
Exercises
The Swiss National Forest Inventory
Estimating change and growth
Exercises
Transect Sampling
Generalities
IUR transect sampling
PPL transect sampling
Transects with fixed length
Buffon’s needle problem
Exercises
APPENDIX A: Simulations
Preliminaries
Simple random sampling
Systematic cluster sampling
Two-phase simple systematic sampling
Figures
APPENDIX B: Conditional expectations and variances
APPENDIX C: Solutions to selected exercises
Bibliography
INDEX