AAIR Project No PL 94-2392
TASK 1: PROGRAMME PLANNING
Participant 05.
Macaulay Land Use Research Institute,
Craigiebuckler, Aberdeen,
Scotland.
AB9 2QJ.
Abstract
This Task has required the participation of all project members in defining the project areas they are working upon in a more systematic manner than that set out in the Technical Annex of the Project Proposal. This Task has identified what is required for the modelling work within the Project. In particular, the data and software functionality that is required for the geographic modelling. An integral part of the Task has been to provide a mechanism for recording an initial assessment of the accuracy of the Tasks' products and the reliability of the input data to each Task.
The outcome of the discussions of the Planning meeting and the subsequent completed record forms, have been analyzed to identify the directions of flow of information and the types of information transfer. Therefore, potential problems in the transfer of data have been identified and can be addressed. Analogous problems associated with the linking of models could not be fully identified and will have to be dealt with in separate Tasks. Any issues that appear to be unresolved have been recorded for subsequent discussion.
The Programme Planning task is designed to coordinate the project objectives by following a planning methodology which explicitly addresses the questions of information production, accuracy and linkage between tasks. The outcome of the Task l is a Reference document for use during the project which contains information on the requirements of each of the Sub-Tasks. This is to assist in coupling together the work and products of each Participant and provide some guidance on the accuracy of the products, the scale, resolution or accuracy of the data required in the production of each Deliverable and one means of tracking sources and magnitudes of error in the modelling.
The results of the Task are presented based upon information available at 30th June 1995. In the course of the project, there may be changes in the software, hardware and data transfer options of some of the participants. Therefore, the summaries of the Tasks and the resources available will be updated as information becomes available. The Task Report provides a documented basis for discussing issues relating to Task linkage and coupling the different models into a single framework within Task 6.
Part 1. Introduction to GIS Planning study and report. Outlines what it is all about. The use of Product Descriptions as planning tool, audit tool, and management tool. Generic product descriptions. Introduce the notion that the planning process (as currently configured/constituted) needs modification for user needs assessment for GIS in a research environment.
Process and scope of studyThis analysis is based on information supplied by project organisations during 1995 and from a project planning workshop held betweeen 23-25th February 1995 at Centro Nacional Informacao Geografica (CNIG) in Lisbon, Portugal. The workshop was itself a fundamental component of this planning study which should properly be considered as a process. The results presented here are, therfore, a summary of the main points established during the workshop; the discussion at the workshop is as important as the completed information products and the benefits from this will be manifest during the course of carrying out the project tasks. For the duration of the Task the Chairman was Dr Richard Aspinall of MLURI. Rapporteur and minute secretary were Dr David Miller and Mr Christopher Quine respectively. The documentation from the meeting plus the forms completed by the Participants comprise the materials for this Task.
The objectives of this report are to:
The planning process is based on description and analysis of the types of information required by individuals and groups; each description is presented as an Information Product.
An information product contains :
i) a sample portion, to scale and fully annotated, of any maps that are required;
ii) a specimen portion of any list or tabular output required showing captions, headings and typical results in maximum field widths needed to present the output;
iii) a list of the datasets needed to make the information product;
iv) a detailed description, using standard, generic terms, of GIS and other functions which are applied to the datasets to make the information product;
v) a list of linkages needed between the datasets in the database;
viii) estimates of tolerance of the information product to different types of error; and
An information product description is written by the individual who requires to use the product. The information product is a tool which serves as both a planning and management tool.
The primary use of an Information Product description is part of this project planning process. A set of information product descriptions collectively define the system capacity, functions and data needs (hardware, software, and data) required and indicate options for its establishment and management (operation and maintenance).
The information product description also has a role as a management tool. This benefits individuals and the group. The information product description, through its structured expression of intended analysis, provides a documented account of the stages involved in information product generation and thereby helps with directed and efficient interactions. The information product also eases communication by providing a documented account of intended use in a common format. Project management should be improved.
The generation of product descriptions for this study also served an eductional purpose introducing individuals to the intended work of colleagues and associated tasks at an early stage of the research investigation.
Information Product
...a formal description of outputs
required from analysis with GIS or other modelling or analytical
software,datasets needed, functions used, frequency of production, database
linkages needed, error tolerances, and costs and benefits
The February meeting produced Task linkage charts (which are presented in Appendix A) which indicate the direction and nature of the flow between participants in each Task. A tabulation of the exchange formats between Participants in each Task is presented in Appendix B. The links are identified in terms of either the Deliverable that is transferred or information relating to a Sub-Task upon which Participants are collaborating. These charts will be available on the WWW and ftp sites as described under Task 8.1. In certain cases the exact nature of the link between participants within the same subtask has not yet been defined.
The second component of the Task has been the completion of the Project Description forms (presented in Appendix C). Most of the Sub-Tasks have had at least one form completed. The results from these forms indicate the exact product that is required, the data and models that will be used or needed and the desired or required accuracy, scale or resolution of the data.
From those forms that were completed it has been possible to identify which formats or media will be required for each data transfer between Participants. The most commonly available GIS package is Arc/Info; the database packages are Oracle and dBase; numerical modelling is undertaken across a broader range of packages and further discussion will be undertaken to identify any problems that this may pose. Computer hardware usage is PC, Sun and Integraph facilities of which PCs are present is every Participating group and direct or indirect access to a Sun (Unix or Solaris) facility is available to every Participant. Data transfer is most commonly:
ASCII files using "ftp" for Tasks 1 to 4 and 8;
Arc/Info export format (Version 7) for map data in Tasks 5 to 8;
Erdas 7.5 for raster data (Digital Elevation Models or satellite imagery) from Tasks 4 to 8;
Word Perfect or Word for report documents. Diagrammatic software has still to be assessed;
Electronic transfer is the most widely employed media and means.
These formats and means of transfer may change in the duration of the project.
Considerable discussion will be required with respect to the use of programming languages and modelling/statistical packages. The most widely used programming language is FORTRAN but this is not appropriate for all models and interface formats between packages will require to be standardized (eg. between NRS and Joensuu).
The use of macros and scripts (such as those for Arc/Info) are likely to be widely used in Task 6. Such scripts should be transferrable to all partners with possible exceptions of AU and UG until ArcView 2 is available to those Participants.
Model development will probably be most effectively undertaken on the software of the host Participant, for their own Tasks. Relatively little model development work is shared between participants and sites. The principal exceptions are the tree breakage model between NRS, Joensuu and Umea, in which case a certain amount of transfer of data files may be facilitated across the internet for testing on the software at each site. Further discussions will be required to identify whether there is an issue relating to running models at a central location.
Further work in Task 1.2 will refine the nature of the sources of uncertainty in the data and models and the implications of such uncertainty on the products.
Task 1 Project Planning
Task 1.1 Planning design
The details of this Task are outlined above.
Deliverables:
1. Report (this document);
2. Product Description (Appendix C);
3. Documentation on standards and protocols;
Links:
To all Tasks in the project.
General Inputs. Proceedings of Planning meeting in February 1995, Lisbon (documentary) and contents of the Programme Planning forms (Appendix C).
General Outputs. Documentation of Tasks, their data and model requirements , their outputs and linkages with other Tasks. Identification of issues to be addressed relating to data, models or presentation.
Data transfer in Word Perfect or ASCII, via ftp to all Participants.
Issues: Not each Sub-Task has a completed Product Description. certain conclusions may require revising in light of later submissions of Product Descriptions. In addition, the technical basis for some of the Sub-Tasks may alter during the time of the project and consequently, the conclusions in this report may be revised during the following 30 months.
Task 1.2 Scale
Deliverables:
4. Guidance on levels of certainty.
Links:
To all Tasks in the project.
General Input. Documented estimates of the scale and accuracy of the output of each Task and the Deliverables.
General Output. Recommendations on the scale or resolution of data to achieve Deliverable. (This sub-Task runs in parallel with Task 6).
Data transfer in Word Perfect or ASCII, via ftp to all Participants.
Issues: Not each Sub-Task has a completed Product Description. Further guideline information will be drawn up in the course of the project. Links between Tasks were expressed in terms of Deliverables in Appendix B of the Task Report. Some additional links may be identified (or existing ones terminated) in due course.
Task 2. Quantification of Component Factors Controlling Snow, Wind and Fire Damage.
Task 2.1 : Wood Property Factors Controlling Wind Damage
Deliverables:
1. Wood strength properties in relation to wind.
Links:
To Tasks 3.1, 3.3 and 4.2.
General Input. Field measured variables of trees, including height, diameter, wood density, compression, slope of the grain and crown diameter.
General Output. Tabular and statistical data on the probability of trees standing, snapping or uprooting and correlations between stem and root strength and turning moment.
Data transfer to later Tasks using ASCII or Excel spreadsheets.
Issues: Some data may be input to PC Arc/Info for assessment of the spatial distribution of the trees that were sampled. Locational accuracy of trees that are sampled should be approximately +- 2.5m on a local reference system and +- 3m on GB National Grid.
Task 2.2 : Wood Property Factors Controlling Snow Damage
Deliverables:
1. Wood strength properties in relation to snow damage.
Links:
To Tasks 5.1, 6.1 and 6.2
General Input. Data from an existing database of tree sample data.
General Output. Values for parameters which control snow damage.
Data transfer is by ASCII files.
Issues: Is Digital Eevation Model available for use in climatological work in task 5.1? What is the time period over which the climate data is available, compared to the data from observations of the sample plots?
Task 2.3 : Tree and Soil Factors Controlling Wind Damage
Deliverables:
2. Database of all available tree pulling experiments in agreed format.
3. Method for applying results of static pulling experiments to the understanding of damage due to dynamic loading.
4. Method for accounting for the effects of root fatigue following repetitive loading.
Links:
To Tasks 3.1, 3.3, 4.1, 4.2, 5.1, 5.2 and 6.1
General Input. Existing database of tree pulling data and new field observations of tree pulling.
General Output. Enhanced database of tree pulling data, turning moments and algorithms.
Data transfer will be by ASCII files, spreadsheets (still to be decided) and computer code (FORTRAN or C++, or Pascal).
Issues: Computer code still to be decided and spreadsheets to be agreed upon between Joensuu and NRS. Are locations of tree pulling experiments recorded. If so, to what level of accuracy?
Task 2.4 : Vegetation Characteristics Controlling Fire Damage
Deliverables:
5. Pure spectral signature data set for fire risk.
Links:
To Tasks 3.4, 4.3, 5.4, 5.5 and 5.6
General Input. Sampled fuel material from the field plus locations for subsequent laboratory analysis.
General Output. Spectral signatures (both pure signatures and mixed signatures) for different fuel types.
Data transfer is ASCII file.
Issues: Accuracy of the GPS locations in woodland environment.
Task 3. Treelevel Models to Predict Circumstances for Damage
Task 3.1 : Empirical Models of Breakage and Overturning of Single Trees by Wind
Deliverables:
Links:
From Tasks 2.1 and 2.3. To Tasks 3.3, 4.1, 4.2 and 6.2
General Input. Parameters relating to individual trees, including wood strength, height and diameter.
General Output. Empirical model of breakage and overturning of a single tree - site based. Critical windspeed for breakage and overturning.
Data transfer of ASCII files, Excel spreadsheet and model code in Mathcad 5.0+ or Turbo Pascal 7.0.
Issues: Relationship between NRS and Joensuu programming language and modelling software to be resolved. Possible exchange of program code or exchange via internal data format. Data from Tasks 2.1 and 2.3 in ASCII and statistical models in Excel. In what form are the uncertainty measures provided for Task 6.2.
Task 3.2 : Empirical Models of Breakage of Single Trees by Snow
Deliverables:
Links:
General Input:
General Output:
Data transfer
Issues:
Task 3.3 : Mechanistic Model for Wind and Snow Damage of Single Trees
Deliverables:
Links:
From Tasks 2.1, 2.3 and 3.1. To Tasks 4.1, 4.2, 5.1 and 6.1.
General Input. Existing mechanistic model and additional data from Finland, Sweden and the United Kingdom.
General Output. An improved mechanistic model for wind and snow damage of a single tree (FORTRAN code, possibly re-written into C++).
Data transfer will be ASCII files, Excel spreadsheets , FORTRAN code and other code not yet specified. Data from Task 2.1 in ASCII and statistical models in Excel.
Issues: Computer language to be used for linking Joensuu and NRS programs., exchange of code or transfer of results via intermediate format. Is locational information required from the tree pulling experiments?
Task 3.4 : Leaf Area Index (LAI) Estimation Subtask 2.1 : Wood Property Factors Controlling Wind Damage
Deliverables:
Links:
From Task 2.4. To Tasks 4.3, 4.4, 5.5, 5.6 and 6.1.
General Input. Destructive LAI measurements at the tree level and estimations of LAI at plot and stand levels.
General Output. Models for LAI estimation using biometric data.
Data transfer will be ASCII files.
Issues: Will the spectral data be directly input to an image processing software package? Are locations of tree required for later reference, particularly if there are simulataneous observations on ground and helicopter (or aircraft)? How is the sampled fuel data from Task 2.4 combined with the tree level observations?
Task 4. Defining Standlevel Variability to Permit Application of Treelevel Models
Task 4.1 : Wind Flow across and within Stands using Airflow Models
Deliverables:
1. Methods for incorporating and variability into overall model and to modify relationships developed for single trees.
Links:
From Tasks 2.3, 3.1, 3.2 and 3.3. To Tasks 5.1 and Task 6.
General Input. Geographic measures of tree stands, such as distance from stand edge, tree spacing and tree height and ninformation derived from the tree pulling database. Mechanistic model for wind and snow damage of single trees.
General Output Relationships for wind-loading with respect to position within the stand.
Data transfer will be numerical values in ASCII format and FORTRAN computer code and in format of the tree pulling database.
Issues: The scale and time of stand observations (eg. from tree pulling database) to match with the scale of relevance of the airflow model. Matching scales and temporal resolution of Umea and NRS models. In what form will the empirical model of breakage be represented for use with airflow models? How reliable are the estimates of stand level variability?
Task 4.2 : Variation in Measured Tree Characteristics within Stands using Various Methods Including Remote Sensing
Deliverables:
2. A method for describing variation in tree heights across stands.
Links:
From Tasks 2.1, 2.3, 3.1, 3.2 and 3.3. To Tasks 5.2 and Task 6.
General Input. Literature review of available information relating to variability in such as tree heights, diameter, volume plus tree pulling database. Measured variation in tree height using aerial photographs.
General Output. Literature-based guidelines complementing 3D model of tree stands for scaling up single tree models to forest level.
Data transfer is by ASCII files and image formats (Erdas/IDRISi). Locational data from Task 2.1 in ASCII file or Arc/Info export format at a resolution better than 5 m and an accuracy of +-2.5m.
Issues: Suitable ground control accuracy in x, y and x to rectify aerial photographs and create Digital Elevation Model of sufficient accuracy. Appropriate scales of aerial photographs to height of trees and control point accuracies. Is locational data recorded in the tree pulling database? How accurately is it recorded? Can the empirical model of breakage be represented as a raster image? How will the reliability of the orthophotographs and the digital elevation models be represented (in space and time)?
Task 4.3 : Plotlevel Mixture Modelling
Deliverables:
3. A method for spectral unmixing of tree canopy and understorey spectro-radiometer signals.
Links:
From Tasks 2.4 and 3.4. To Tasks 5.4, 5.5 and 5.6.
General Input. Field materials gathered in Task 2.4 and LAI measures from Task 3.4.
General Output. A method for spectral unmixing of tree canopy and understorey.
Data transfer by ASCII files.
Issues: Reliability of spectral unmixing which matches spatial and spectral resolutions of available satellite imagery. Sensitivity of spectral unmatching to characteristics of satellite image. What role does the LAI play in the spectral unmixing?
Task 4.4 : Preparation of Knowledge Base for Fuel Hazard Mapping
Deliverables:
Links:
From Tasks 2.4 and 3.4. To Tasks 5.5, 5.6 and Task 6.
General Input. Literature review of relevant information. Knowledge engineering of experts.
General Output. Knowledge-base for fuel hazard mapping.
Data transfer may be by ASCII files, image files (Erdas, Intergraph), digital map data (Arc/Info, Intergraph), third party code and "knowledge transfer" techniques.
Issues: Coupling of knowledge-base with GIS and image processing. Therefore, interfacing knowledge-base package with a GIS; selecting levels of close or loose coupling; identifying host computer for support of suitable software. Linkage of knowledge-base to framework in Task 6. Will the knowledge-base system include all LAI data (eg. as look-up tables)?
Task 5. Regionallevel Snow, Wind and Fire Risk Models
Task 5.1 Climatology of Wind and Snow in Relation to Topographic Variability and Temporal Incidence
Deliverables:
1. A working model for predicting wind and snow at regional levels
2. Error estimates for wind and snow predictions.
8. Regionallevel snow and wind maps ?
Links:
From Tasks 2.2, 2.3 and 5.2. To Tasks 5.2 and 6.1.
General Input. Records from Meteorological stations, wind atlas and combined with topographic data.
General Output. A method for deriving the climatology of wind and snow. Surfaces representing snow lie for test sites and associated error estimates. Predicted windiness at two test sites.
Data transfer will be as image data (Erdas, IDRISi).
Issues: Scale and resolution of the terrain data for the two UK test sites to be appropriate to match with the field measurements. Further discussion to take place with respect to resolution of output versus modelling requirements. Relationship between snowfall data and predicted windiness to be addressed, timlieness of snowfall data compared to windiness data. Relate WASP model and test site data at appropriate scales. Check DEM for Sweden and its appropriateness for use in the climate modelling. How will surface roughness be measured? Is resolution and magnitude of variability for stand level (in Task 4.1) for the same sites as Task 5.1?
Task 5.2 Construction of Geographical database and Use of Image Processing
Deliverables:
3. Spatial datasets and geographic database for each site.
4. Satellite imagery prepared for use in Task 6.
Links:
From Tasks 2.3, 4.2, 5.1. and 5.3. To Tasks 5.1, 5.4, 5.5, 5.6, Task 6 and Task 7.
General Input. Digitizing of graphical map originals as input to digital data layers to the spatial database, for each test site. Also rectification of satellite imagery or aerial photography, construction of database entries with location attributes and purchase of third party digital geographic data, for each test site and region of interest.
General Output. GIS data layers for each test site and region of interest.
Data transfer will be by ASCII file, image data (Erdas and IDRISi) and digital map file (Arc/Info export) format.
Issues: Requirements for records of data compilation for inclusion in meta-data listings for subsequent use in Tasks. Clarificaton of copyright and rights of access to data at different sites or from different sources. Linkage of the tree pulling database to GIS: what attributes are to be chosen for linkage to models? What attributes are required to link the knowledge-base in Task 5.5 to a GIS?
Task 5.3 Explore Relevance of PanEuropean Datasets of Eurostat and CORINE
Deliverables:
Links:
From Task 1. To Tasks 5.2 and Tasks 6, 7 and 8.
General Input. Access requirements are confined to a limited number of on-line databases, datasets and printed documentation.
General Output. is a text file (Word Perfect 5.1) and an ASCII version for possible inclusion into WWW pages.
Data transfer is by ASCII files.
Issues: Matching available data with that required for developing and validating models. Suitability of data for geographic areas outside the geographic area of data compilation. Options available for interface for Task 6, such as RIVM land use database interface for PCs.
Task 5.4 : Regional Classification of Fuel Types Using a Mixture Models and Functional Vegetation Classification
Deliverables:
5. Fuel type maps.
6. Maps of vegetation structure for test sites.
Links:
From Tasks 2.4, 3.3, 4.3, 4.4 and 5.2.To Tasks 5.2, 5.5 and 5.6.
General Input. Satellite imagery, laboratory analysis of spectral data. and plot level mixture models.
General Output. Functional/structural vegetation classification of fuel types.
Data transfer by ASCII file and (possibly) image data by Erdas and Intergraph.
Issues: To what extent does the time of year and ambient conditions affect the relationship between the results of the laboratory analysis and the satellite imagery? Will the plot level mixture modelling be used in the classification of fuel types? How will the reliability of the results be represented?
Task 5.5 : Knowledge Base System Construction for Fuel Hazard Mapping
Deliverables:
Links:
From Tasks 2.4, 3.4, 4.3, 4.4 and 5.2. To Tasks 5.6 and Task 6.
General Input. Knowledge-base from Task 4.4 and instance data from imagery and map origins. LAI data from Task 3.4.
General Output. Knowledge-base shell and fuel hazard map.
Data transfer by ASCII files, rule modules from knowledge-base in form to be decided, images in Erdas or Intergraph and digital map data in Arc/Info export format or Intergraph.
Issues: Formalization of knowledge-base from Task 4.4. Formalizing expression of instances of fuel hazard and transfer of data between knowledge-base and GIS/image processing system (eg. Task 5.2). Linkage of KBS to overall framework in Task 6. Description of meta-data for the plot level mixture model.
Task 5.6 : Fire Risk Model Construction
Deliverables:
7. Fuel hazard maps.
8. Regional level fire risk model.
Links:
From Tasks 2.4, 3.4, 4.3, 4.4, 5.2, 5.4 and 5.5. To 5.2 and Task 6.
General Input. Fuel hazard map and airflow model.
General Output. Estimate of spatial distribution of fire spread.
Data transfer by imagery (Erdas or Intergraph), digital map data by Arc/Info or Intergraph.
Issues: Reliability of airflow model developed in UK when applied in Portugal. Scale and resolution of airflow model from UK test sites compared to scale and resolution of fuel maps in Portuguese test sites.
Task 6. Integrating Components
Task 6.1 : Integrating the Component Models
Deliverables:
Framework for integrating component models
Links:
Links to all Tasks in the project.
General Input. Geographic data overview of Europe, information on the data availability for each country, information on the data used in model development and for validation. Statistical, geographical and rule-based models (housed at MLURI or remaining resident at other Participants') and data for developing or validating models. Information on data and model reliability. mechanistic model of wind and snow damage. Knowldge-base system of fuel types an fire hazard. Information on the linkages between geographic data and models.
General Output. a framework, computer-based, for a user to access models, data (or both) directly for predicting damage and obtaining information on silvicultural strategies to alleviate or ameliorate such damage. Output is graphical, tabular or textual.
Data transfer will be electronic for remote access, use and provision of the results; tabular (ASCII or a report file) in response to the query; graphical (in one of the data formats that will be supported) for subsequent electronic transfer.
Issues: Decisions have still to be reached on whether data and models will all be housed in (at least) one location, or whether the data will reside in remote locations and the models will be central, or whether all the models and data will be remotely situated and access to each site will be through a central point of contact which only manages the nature of the initial enquiry and directs the user to the appropriate location. In what form is the mechanistic model presented? What components of the knowledge-base system will be explicitly represented in the framework (eg. as meta-data)? What level of information will stand level variability be represented?
Task 6.2 : Scale Issues and Error Propagation
Deliverables:
Guidelines on the validity of the scale and resolution of the data for use in different models and the error propagation to be expected from use of certain models.
Links:
Links are to all Tasks in the project.
General Input. Meta-data gathered during Task 5.2 plus information provided in Product Descriptions in Task 1.
General Output. Rules on the likely validity of the use of each model with each dataset. Further guidelines on the relevance of models applied across larger areas of the EC.
Data presentation and transfer will be textual, tabular, or graphical and in ASCII, image or mapped form.
Issues: The exact nature of the scale and error data has not yet been ascertained. This will be completed and assessed as each Sub-Task is completed and the success of the modelling reported upon. How will reliability of stand level variability be assessed and used? How will the reliability measures of the mechanistic model of tree breakage and turnover be represented?
Task 7. Testing Models against Independent Data
Task 7.1 : Model Testing by Application of Change Detection System to Finnish Test Area, Detecting Wind Damage on Stand Level by Landsat TM Images
Deliverables:
1. Statement of success of models
2. Recommendations for improving the models.
3. Limitations of the extent of validity of models.
4. Example of damage minimizing strategy.
Links:
From all Tasks. To Task 8.
General Input. Landsat TM imagery and an existing system for detecting forest change to derive areas of wind and snow damage. Field data for validation of areas assessed to have changed due to damage.
General Output. Assessment of areas of damage and cause of damage. Statement of success and the limitations.
Data transfer will use the protocols developed in Task 6 and between Tasks by the end of Year 2.
Issues: Reliability of change detection system compared to magnitude of changes. Difficulty in ensuring both wind and snow damage plus areas of no damage for adequacy of testing. Further discussion required on populating models and framework with silvicultural `knowledge' to produce damage minimizing strategy. Further discussion required on the means of presentation of damage and minimizing strategies in Task 6.
Task 7.2 : Model Testing through Portuguese Forest Service
Deliverables:
1. Statement of success of models
2. Recommendations for improving the models.
3. Limitations of the extent of validity of models.
4. Example of damage minimizing strategy.
Links:
From all Tasks. To Task 8.
General Input. Weather and fire records, fuel hazard assessment and fire spread model on an independant dataset in Portugal.
General Output. Statement of success of models, limitations of modelling and example of damage minimizing strategy.
Data transfer will use the protocols developed in Task 6 and between Tasks by the end of Year 2.
Issues: Further discussion on whether models will be tested outside Portugal (eg. on UK test site data) and means of presentation of damage and minimizing strategies in Task 6.
Task 7.3 : Model Testing through Forestry Commission Windthrow Monitoring Areas and Forest Districts
Deliverables:
1. Statement of success of models
2. Recommendations for improving the models.
3. Limitations of the extent of validity of models.
4. Example of damage minimizing strategy.
Links:
From all Tasks. To Task 8.
General Input. Data for UK monitoring areas from Task 5.2 plus models integrated in Task 6. Independant validation data from monitoring sites.
General Output. Probability of wind damage in map form and simulations predicted for future years.
Data transfer will be in the form of paper maps, imagery and tabular data for forests use the protocols developed in Task 6 and between Tasks by the end of Year 2.
Issues: Further discussion required on which sites will be kept for validation. Further discussion on manner of integrating rules relating current silvicultural techniques and management style to extent and location of damage.
Task 8. Final Products
Task 8.1 Dissemination of Project Information and Results
Deliverables:
Final reports.
Links:
To all Tasks and participants in the project.
General Input. Textual, diagrammatic and statistical information from each Task. Software and models from each Task.
General Output. Listserver and World-Wide Web (WWW) site will be on a Unix facility at MLURI. Access to both will require connections to electronic communications (Email). Use of the WWW pages will require suitable software, such as Mosaic or Netscape, both of which are available by anonymous file transfer protocol (FTP).
Issues: Several questions remain to be addressed, including copyright, confidentiality and rights of access to software and data produced in the course of the project.
Task 8.2 Dissemination of Project Achievements
Deliverables:
Progress reports, final reports and publicity materials.
Links:
Links are to all Participants, the EC and outside interests.
General Input. Textual, diagrammatic and statistical information from each Task. Software and models from each Task.
General Output. Written materials to EC and scientific journals, meetings and professional reports.
Data transfer is via paper documents and a restricted amount of electronic communications.
Issues: Copyright, authorship and timescales of publications. Future development of the project content.
Literature or Database contents
Task 1.2 Programme Planning questionnaire form replies.
Task 2.2 Database of tree samples at Umea.
Task 2.3 Tree pulling database held at NRS.
Task 5.3. Literature search on geographic data availability and access.
Task 3.3. Development of mechanistic model using existing data and model.
Task 4.1. Development of air flow model across stands using existing data.
Task 4.3. Mixture modelling using data collected in Task 2.3.
Task 4.4. Compilation of knowledge-base for fuel hazard mapping.
Task 5.1. Climatology of wind and snow using existing database data.
Task 5.5. Construct knowledge-base for fuel hazard mapping.
Task 6.2. Scale and error propagation using meta-data.
Task 7.3. Compare predicted and actual damage within windthrow monitoring areas.
Field Data
Task 1.1. Programme Planning questionnaires to Participants.
Task 1.2. Results from data or model analysis.
Task 2.1. Discs cut from sample plots.
Task 2.3, Tree pulling by Joensuu.
Task 2.4. Sampling fuel materials, location (using GPS) and spectral analysis in laboratory.
Task 3.3. Development of mechanistic model using field data from 2.3.
Task 3.4. Leaf Area Index (LAI) and canopy observations.
Task 4.2. Field observations on tree heights for validation purposes.
Task 7.1. Observations of tree damage in Finnish test area for validation.
Land Cover Data
Task 5.2. Construction of geographical database.
Task 6.1. Integration of component models (interface).
Digital Elevation Data
Task 5.1. Construction of geographical database.
Task 5.6. Integration of component models.
Soils
Task 3.1. Empirical model of tree breakage and overturning - site based.
Task 5.2. Construction of geographical database.
Task 6.2. Integration of component models.
Climate
Task 5.1. Map and database information for derivation of maps of windspeed, snow occurrence and snow depth.
Task 5.6. Construction of fire risk model using airflow model from task 5.1.
Task 6.1. Integration of component models.
Task 7.2. Validation data for fire risk model.
Remotely Sensed data
Task 4.2. Aerial photography at 1:10000, 1:15000 and 1:24000.
Task 5.2. Construction of geographical database.
Task 5.4. Regional classification of fuel types.
Task 6.1. Integration of component models.
Task 7.1. Existing Thematic Mapper imagery for validating wind damage in Finnish test area.
The outcome from Task 1 has been a formalization of the direction and type of transfer of data, models or results between participants.
The interchange of deliverables between participants has been identified and the form of the deliverable has been more clearly defined.
The first six months of the duration of the project at MLURI has prepared the basis for the subsequent work on modelling and interfacing between the user, the models and the data. The Programme Planning Task has helped to focus on issues of data and model requirements, specifically, on the extent to which the data is reliable and whether or not it is available. The Data Report of Task 5.3, gives an indication of the nature, extent and (in a limited number of cases) the reliability of the data that are available for Europe-wide and the countries of participating organizations.
The lack of relevant information on the reliability of data may prove a limitation on some of the geographic modelling that will be carried out in later Tasks. It also points to a gap in the descriptions of much of the data that is available, a point that will require addressing with respect to the products of the current project.
Finally, the preparations for the operational use of electronic communications have set in place the infrastructure for exchanging information, whether that be discursive, data or project Deliverables.
The Task Report provides a basis for discussing specific aspects of the coupling together of data and models from each of the Tasks into a single framework for assessing the risk of forest damage due to physical factors.
APPENDIX A
Sub-Task LinkagesAPPENDIX B
Sub-Task Data Exchange
APPENDIX C
Sub-Task Product DescriptionsAPPENDIX D
Data types, software, modelling and transfer media
1. Use of Digital soils data or Inventory.
2. Use of Digital Elevation data.
3. Use of Remotely Sensed Data.
4. GIS and Image Procesing Software.
5. Use of Modelling software.
6. Use of Programming Languages.
7. Digital data transfer media.
8. Word processing software.
APPENDIX E
University of Joensuu:
Participant 01 - Project Coordinator
Professor Seppo Kellomaki,
Dr Heli Peltola,
Ari Talkkari,
Jari Varjo
Faculty of Forestry,
University of Joensuu,
PO Box 111,
Fin80101,
Joensuu,
Finland.
Tel: +358 73 151 3630
Fax: +358 73 151 3590
Email: kello@joyl.joensuu.fi
hpeltola@gis.joensuu.fi
and talkkari@gis.joensuu.fi
ftp site: gis.joensuu.fi
WWW site: http://ix3866.joensuu.fi/~talkkari/homepage.html
Aberdeen University:
Participant 02
Dr Andrew Cameron
Dr Roger Dunham
Department of Forestry,
University of Aberdeen,
St Machar Drive,
Old Aberdeen,
Aberdeen.
AB9 2UD
United Kingdom.
Tel: +44 1224 272677
Fax: +44 1224 272685
Email:for148@aberdeen.ac.uk
or Alistair Law@aberdeen.ac.uk
WWW site: http://www.abdn.ac.uk/~for148
CNIG:
Participant 03
Dr Maria J Vasconcelos
Centro National de Informacoa Geografica,
Rua Braamcamp 82,
Lisbon,
Portugal.
Tel: +351 1 386 0011
Fax: +351 1 386 2877
Email: maria@helios.cnig.pt
and mario@helios.cnig.pt
NRS:
Participant 04
Mr C P Quine
Dr B Gardiner
The Forestry Authority,
Northern Research Station,
Roslin,
Midlothian.
EH25 9SY.
United Kingdom.
Tel: +44 131 445 2176
Fax: +44 131 445 5124
Direct line: +44 131 445 6948
Email:M.A.X.Gust@ed.ac.uk
Ministry of Agriculture:
Participant 06
Dr Manuela Baptista
Instutio Florestal,
Ministerio da Agricultura,
Av. Joao Critismo,
26 28 1000 Lisbon,
Portugal.
Tel: +351 1 356 1894
Fax: +351 1 541 462
Email:
SLU:
Participant 07
Dr Erik Valinger
Faculty of Forestry,
Swedish University of Agricultural Sciences,
Umea,
Sweden.
Tel: +46 90 166 284
Fax: +46 90 166 817
Email: ronja@nana.slua.se
Galway:
Participant 08
Dr Michael Rodgers
Brian Flynn
Department of Civil Engineering.
University College Galway,
Galway,
Ireland.
Tel: +353 91 244 11
Fax: +353 91 750 507
Email: brian.flynn@ucg.ie