From lidar scans to roughness maps for wind resource modelling in forested areas

Rogier Ralph Floors*, Peter Enevoldsen, Neil Davis, Johan Arnqvist, Ebba Dellwik

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

382 Downloads (Pure)

Abstract

Applying erroneous roughness lengths can have a large impact on the estimated performance of wind turbines, particularly in forested areas. In this study, a new method called the objective roughness approach (ORA), which converts tree height maps created using airborne lidar scans to roughness maps suitable for wind modelling, is evaluated via cross predictions among different anemometers at a complex forested site with seven tall meteorological masts using the Wind Atlas Analysis and Application Program (WAsP). The cross predictions were made using ORA maps created at four spatial resolutions and from four freely available roughness maps based on land use classifications. The validation showed that the use of ORA maps resulted in a closer agreement with observational data for all investigated resolutions compared to the land use maps. Further, when using the ORA maps, the risk of making large errors (> 25 %) in predicted power density was reduced by 40–50 % compared to satellite-based products with the same resolution. The results could be further improved for high-resolution ORA maps by adding the displacement height. The improvements when using the ORA maps were both due to a higher roughness length and due to the higher resolution.
Original languageEnglish
JournalWind Energy Science
Volume3
Issue number1
Pages (from-to)353-370
Number of pages18
ISSN2366-7443
DOIs
Publication statusPublished - 2018

Bibliographical note

© Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License

Cite this

@article{cbe83d98accc48e1b76b23b5cc8c5aa1,
title = "From lidar scans to roughness maps for wind resource modelling in forested areas",
abstract = "Applying erroneous roughness lengths can have a large impact on the estimated performance of wind turbines, particularly in forested areas. In this study, a new method called the objective roughness approach (ORA), which converts tree height maps created using airborne lidar scans to roughness maps suitable for wind modelling, is evaluated via cross predictions among different anemometers at a complex forested site with seven tall meteorological masts using the Wind Atlas Analysis and Application Program (WAsP). The cross predictions were made using ORA maps created at four spatial resolutions and from four freely available roughness maps based on land use classifications. The validation showed that the use of ORA maps resulted in a closer agreement with observational data for all investigated resolutions compared to the land use maps. Further, when using the ORA maps, the risk of making large errors (> 25 {\%}) in predicted power density was reduced by 40–50 {\%} compared to satellite-based products with the same resolution. The results could be further improved for high-resolution ORA maps by adding the displacement height. The improvements when using the ORA maps were both due to a higher roughness length and due to the higher resolution.",
author = "Floors, {Rogier Ralph} and Peter Enevoldsen and Neil Davis and Johan Arnqvist and Ebba Dellwik",
note = "{\circledC} Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License",
year = "2018",
doi = "10.5194/wes-3-353-2018",
language = "English",
volume = "3",
pages = "353--370",
journal = "Wind Energy Science",
issn = "2366-7443",
publisher = "Copernicus GmbH",
number = "1",

}

From lidar scans to roughness maps for wind resource modelling in forested areas. / Floors, Rogier Ralph; Enevoldsen, Peter; Davis, Neil; Arnqvist, Johan; Dellwik, Ebba.

In: Wind Energy Science, Vol. 3, No. 1, 2018, p. 353-370.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - From lidar scans to roughness maps for wind resource modelling in forested areas

AU - Floors, Rogier Ralph

AU - Enevoldsen, Peter

AU - Davis, Neil

AU - Arnqvist, Johan

AU - Dellwik, Ebba

N1 - © Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License

PY - 2018

Y1 - 2018

N2 - Applying erroneous roughness lengths can have a large impact on the estimated performance of wind turbines, particularly in forested areas. In this study, a new method called the objective roughness approach (ORA), which converts tree height maps created using airborne lidar scans to roughness maps suitable for wind modelling, is evaluated via cross predictions among different anemometers at a complex forested site with seven tall meteorological masts using the Wind Atlas Analysis and Application Program (WAsP). The cross predictions were made using ORA maps created at four spatial resolutions and from four freely available roughness maps based on land use classifications. The validation showed that the use of ORA maps resulted in a closer agreement with observational data for all investigated resolutions compared to the land use maps. Further, when using the ORA maps, the risk of making large errors (> 25 %) in predicted power density was reduced by 40–50 % compared to satellite-based products with the same resolution. The results could be further improved for high-resolution ORA maps by adding the displacement height. The improvements when using the ORA maps were both due to a higher roughness length and due to the higher resolution.

AB - Applying erroneous roughness lengths can have a large impact on the estimated performance of wind turbines, particularly in forested areas. In this study, a new method called the objective roughness approach (ORA), which converts tree height maps created using airborne lidar scans to roughness maps suitable for wind modelling, is evaluated via cross predictions among different anemometers at a complex forested site with seven tall meteorological masts using the Wind Atlas Analysis and Application Program (WAsP). The cross predictions were made using ORA maps created at four spatial resolutions and from four freely available roughness maps based on land use classifications. The validation showed that the use of ORA maps resulted in a closer agreement with observational data for all investigated resolutions compared to the land use maps. Further, when using the ORA maps, the risk of making large errors (> 25 %) in predicted power density was reduced by 40–50 % compared to satellite-based products with the same resolution. The results could be further improved for high-resolution ORA maps by adding the displacement height. The improvements when using the ORA maps were both due to a higher roughness length and due to the higher resolution.

U2 - 10.5194/wes-3-353-2018

DO - 10.5194/wes-3-353-2018

M3 - Journal article

VL - 3

SP - 353

EP - 370

JO - Wind Energy Science

JF - Wind Energy Science

SN - 2366-7443

IS - 1

ER -