Forest Biometry, Modelling and Information Sciences FBMIS

1
Forest Biometry, Modelling and Information
Sciences(FBMIS) the Forest Model Archive(FMA)

• Professor Keith Rennolls
• University of Greenwich

2
Stats and Models in Forestry

• Distant history Germany India
• Graphical methods for height-age, volume-age
  Yield Tables
• Statistics The Origins
• Galton, Pearson, Yule
• Fisher 1925 , Neyman-Pearson 1927,
• Yates (Census of Woodlands, 1942)
• etc
• Up to about 1965
• IUFRO 6.02 The Advisory Group of Forest
  Statisticians
• Bertil Matern Spatial Variation 1960
• John Jeffers.
• Experimental design and analysis in forest
  research 1960
• The electronic digital computer in forestry 1961
• Al Stage TRAS growth simulation program 1962
• Prodan Forest Biometrics 1968
• Paul Schmid-Haas Edge Effects 1969
• Van Laar Growth modelling 1969
• Joe Christie Yield Tables The Red book

3
continued

• After1970 The Modern Age.
• John Jeffers, J.N.R. (Ed). 1972. Mathematical
  models in ecology.
• Collier Dawkinss Statforms 1975
• Alder Tropical forest models
• Bill Warren Spatial and Temporal models in
  Forestry
• Oscar Garcia System-Dynamic Stand-Level Forest
  Models
• Anniki Makela Process Based Forest Modelling
• Bob Monserud Boreal and Temperate Forest
  Modelling,
• PROGNOSIS and Forest-Vegetation-Simulator (FVS)
• Jerry VanClay Tropical Forest Management Models
• Any many others
• but currently probably only 200 Forest
  Biometricians,
• and possibly 400 in Mensuration Growth and
  Yield.
• Quite specialized, quite small

4
Me Forest Stats Modelling

• Forestry Commission Statistics Branch Alice
  Holt 1976-1985
• First IUFRO 6.02 Conference 1978 Frieburg
• Warren, Matern, Prodan, Ek,
• About 1980 Forest Modelling Discussion Group,
  FMDG.
• IUFRO 6.02, 4.11, 4.01 Office holder since 1990
• IUFRO DSA (Div. 4) 2005
• EU FIMP Indonesia 1996
• Tropical Forest modelling, inventory, remote
  sensing, biodiversity

5
Forest Biometrics and Modelling Publication
Outlets

• Forest Science
• Canadian Journal of Forest Science
• Forest Ecology and Management
• Remote Sensing of the Environment
• Ecological Modelling
• Tropical Forest Ecology
• Journal of Ecology
• Biometrics
• Commonwealth Forestry Review
• National Forestry Journals e.g. Forestry
• Only ever about 1 in 10 or 20 on Forest
  Biometrics and Modelling.
• IUFRO 6.02, 4.11, 4.03 Conference Proceedings
• Grey Literature
• Black Hole Literature?

6
IUFRO 4.11 Greenwich, 2001 Forest Biometry,
Modelling and Information Science (FBMIS)

• The Call
• Why Information Science(s)?
• 1. Fishers Information Theory
• The Basis of Modern Inferential Statistics
• 2. Shannons InformationTheory
• A popular species diversity measure
• 3. Information Systems
• Storing and managing data and information
• 4 Information Science
• Librarianship concerned with catalogues, keys,
  and restricted vocabularies metadata
  ontologies
• 5. Forest Information Systems
• Combining 3. and 4. GFIS, EFIS, CFIS, NEFIS.

7
The FBMIS Proceedings

• The First IUFRO 4.11 web-proceedings
• Systems for Forest Information Conservation and
  Sharing
• (Chairs Ibrahim and Paivenen)
•  
• Risto Paivinen, Tim Richards, Nguyen Thanh Binh,
  Richard Wood , Margherita Sini (Invited
  paper)The IUFRO Global Forest Information
  Service
• Towards a Common Resource.
• Patrick D. Miles. USDA FS, St. Paul. Information
  Systems for the Forest Resource Analyst.
• Keith Rennolls , Moh Ibrahim and Peter Smith.
  University of Greenwich.A Forest Model Archive?
• Nell Baker and Howard Wright. University of
  OxfordThe UK Archive of Tropical Forest
  Inventory
• (ATROFI) conserving the data.  

8
The Forest Model Archive (FMA)

• The FMA a web portal www.forestmodelarchive.i
  nfo
• The Full FMA Standards Specification
• The resources associated with a fully documented
  model in FMA are expected to include the
  following as the adopted STANDARD
• A description of the theory of the model
• A listing of the code for the model algorithm
• The fitting criterion used
• The fitting algorithm used
• Test data, with metadata
• The results of fitting the model to the test data
  using the fitting algorithm
• A new way to publish models

9
The FMA Aims

• To Conserve Forest Models
• To Enable easy re-use of Forest Models
• Even for non-experts
• Especially for the Developing World
• A latter-day Stat-forms?
• To Foster new forest model development
• Particularly for Tropical Forests
• For Remote-Sensing based Inventories
• To Allow easy comparison of Models
• On standard datasets
• On new datasets
• To Facilitate the building of new Hybrid models
  from previous models
• To Engender international forest modelling
  collaboration

10
FMA Challenges

• Modellers prefer to model rather than re-use
  another modellers model.
• There is no kudos in re-use
• Many of the established BIG Models have their own
  well supported web sites
• FMA building is time-consuming and needs funding
  
• not easy!
• FMA construction does not rate for RAE!
• Only publications, funding and PhDs do.
• An Archive as merely a repository of dead
  models does not capture its potential future
  contribution.
• Something is needed in addition to indicate that
  the FMA is State of the Art

11
FBMIS the e-journal(FBM Information Sciences
!)

• Parents
• Absence of a Specialized Forest Biometrics and
  Modelling publication outlet
• The FBMIS Conference Proceedings
• All the Black-Hole Proceedings of the past
• The FMA
• FBMIS www.fbmis.info

12
FBMIS 2003-4

• OSCAR GARCIADimensionality Reduction in Growth
  Models an example.
• ALMUTH WAMELING JOACHIM SABOROWSKI
  Construction of Local Confidence Intervals for
  Contour Lines.
• GEORGE GERTNER
• Assessment of Computationally Intensive Spatial
  Statistical Methods for Generating Inputs for
  Spatially Explicit Error Budgets.
• ROBERT A. MONSERUDEvaluating Forest Models in a
  Sustainable Forest Management Context.
• Special Invited PaperJOE LANDSBERG
  (Biography)Physiology in forest models history
  and the future
• Special Invited PaperTASITI SUZUKI
  (Biography)Gentan Probability and the Concept of
  the Normal Wood in the Wide Sense
• Special Invited Paper JOHN JEFFERS
  (Biography)Whither biometrics?
• ANDREW P. ROBINSON, AMY L. POCEWICZ, and PAUL E.
  GESSLERA Cautionary note on Scaling Variables
  that appear only in Products in Ordinary Least
  Squares
• NICOLAS PICARD and ALAIN FRANCApproximating
  spatial interactions in a model of forest
  dynamics.
• JEROME K VANCLAYIndicator Groups and Faunal
  Richness.

13
FBMIS Challenges?

• The Challenges for the FBMIS
• E-journals are not recognized as fully archival
  by some authorities
• The main-stream journals have no publication
  charge
• All the main journals are now on-line
• FBMIS needs a funding line currently there is a
  publication charge of 150
• Sponsorship is needed
• Managing and Editing a journal is time-consuming
• There is relatively little RAE credit for doing
  so.

14
FBMIS Future?

• New Developments
• Special Issues for IUFRO 4.01 conferences?
• Web-conferences?
• Index to On-line Biometrician Personal Archives?
• e.g. Oscar Garcia Jerry VanClay
• Future Home, after Greenwich?
• A main-stream publisher?
• Other Institutions or Individuals?
• IUFRO? FAO? CABI?

15
FMA FBMIS final view

• The FMA was the parent of FBMIS
• The FMA is now the brother of FBMIS
• FBMIS as a journal is archival by nature
• The FMA is archival, but not only archival
• The FMA is a publication venue of a multiple
  dimensions, with great creative potential
• Maybe the future should be an FBMIS Portal,
• encompassing both the FBMIS e-journal,
• and the FMA?