AAIR Project: Sub Task: 5.1 Temporal and spatial variability of the wind climate

Participant : The Forestry Authority,

Northern Research Station,

Roslin, Midlothian,

Scotland, EH25 9SY

Contact: C P Quine, B A Gardiner, P D Bell.

Description:

Derivation of method of quantifying temporal and spatial variability in wind climate as input to predictive model of wind damage.

Inputs:

Digital terrain models of study sites in Kintyre and Cowal, West Scotland. Surface roughness map (at time of windspeed measurements) from height and compartment records/maps. Test data from anemometer sites at known locations.

Outputs:

Predicted windiness (relative and absolute) using linear airflow models (MS Micro and WASP) and geographic predictors (derived independently from tatter flags).

Method:

Two study sites have been selected where field measurements have been taken at a range of topographic positions. Actual measurements will be compared with predictions derived by a variety of methods; some work will be required to resolve how to best represent the time dimension eg weighting by average wind rose etc. A blend of the two may be required in complex terrain eg for funnelling and separation. Extreme value methodology (Gumbel and newer variants) will be explored and applied to wind climate predictions derived from spatial analysis.

Uses:

The recommended method will be used to derive probability of windspeed (and combined with other work on snow in this sub-task by MLURI and Umea) within the overall Task 6 predictive model. The method will be applied to test sites in Task 5.2 by Forestry Authority, by CNIG in Task 5.6 to allow validation in Task 7.2 by CNIG and PFS, and in 7.3 by Forestry Authority.

References:

Hannah, P. (1993). Application of wind modelling techniques in complex terrain, PhD thesis University of East Anglia.

Inglis, D.W.F. (1992) A field and modelling study of airflow over complex terrain and in forest canopies. PhD thesis University of Manchester.

Troen, I and Petersen, E.L.(1989) The European Wind Atlas. Riso National Laboratory, Denmark.

Sub Task: 5.1

Participant: The Forestry Authority

MODELS USED FOR DATA ANALYSIS

I. Model Name:WASP;MS3-Micro;Windiness scores

II. Source of Model:Riso, Denmark;AES Canada;NRS

III. Computer language:?;Lahey Fortran;Turbo pascal

IV. Hardware needed: 386 with Maths coprocessor

V. Data Inputs needed:

• Digital Terrain model (gridded for Wasp and MS3)

• Surface roughness (from vegetation height)

• Initialising direction

VI. Model Outputs

• A. Tables/Statistics

• Wasp - wind speed/direction frequency table
  
  

• B. Maps/GIS datalayers

• MS3 - Wind speed multiplier

• Wasp - wind speed multiplier from batch file

• Windiness score - windiness

VII. Accuracy/Sources of Uncertainty/Method of Handling:

• Accuracy to be tested as part of task

• Slope limitations, flow separation and funnelling

• Initialising data from Met Office

• Assumptions of static time-series in historic data

• Assumptions of neutral atmospheric conditions

VIII. Existing linkages between data and software:

• Surfer<>MS3-micro

• OS DTM>IDRISI windiness score>CSS Statistica

• Surfer<>Wasp

IX. GIS Software used:

• IDRISI

X. Modelling Software used:

• MS3-Micro

• Wasp

• Turbo pascal

XI. Database Software used:

• Excel

• CSS Statistica

XII. Any other comments:

Prepared by C P Quine/B A Gardiner/P D Bell

May 1995