Saarblitz, Wind farm in HDR,, creative commons by-nc-sa 2.0


Managed by

Site Description: 

The location of the site can be seen in figure 1, where the red box indicates the area of figure 2 and 3 which shows the elevation and tree height deduced from airborne laser scans. Figure 4 and 5 shows the same thing but in larger zoom. The forest consists predominately of spruce, but is mixed with pine and some deciduous forest types. The site includes a variety of heterogeneities in both topography and land cover. The forest itself is also heterogeneous with patches of trees of different densities, height and age. The site is typical for wind energy exploration in Sweden and offers many possibilities of investigating heterogeneities in the wind field due to vegetation and elevation.

Data Provider: 

Johan Arnqvist.

The data has been collected within a research collaboration that consists of Swedish Vindforsk IV research project ForestWind, EU ERAnet NEWA (New European Wind Atlas) project and industry partner OX2 Vindkompaniet.

Data accesibility: 

The full data set will be open access in the future, but is closed in part due to keeping the benchmark blind and in part to protect the industry partner until development of the wind turbine site.
The validation data will be accessible after the benchmark is closed.
Surface data such as elevation height, forest height, Plant Area Density (PAD) is available to the participants of the benchmark in 10x10 m resolution within an area of 40x40 km surrounding the tower.


Tower measurements
The tower was 180 m high, square, with 1.2 m wide sides. It was equipped with 8 Metek 3D ultra sonics, 7 Thies first class anemometers, 3 Vaisala anemometers, 3 wind vanes, and a temperature profile system. The majority of the instrumentation in the tower was financed by the ForestWind project, apart from the cups and vanes which was financed by OX2 Vindkompaniet. Additional instruments, such as two Kipp & Zonen 4-way radiation sensors at 30 and 0.5 m have been placed at the location of tower as part of the NEWA campaign. The tower instruments were operational between June 2015 and July 2017.

Remote sensing instruments
The location of the remote sensing instruments is best seen in Figure 4-5 where they are indicated by their model name. At the tower site there was a Zephir 300 lidar measuring at the same heights as the tower, but recording raw data in order to validate the lidar performance and provide redundancy to the tower measurements. An additional Zephir 300 was deployed at the most western location, measuring the wind speed up to 300 m and also recording raw data. Most of the validation points consist of AQ systems AQ510 sodar profilers. They measure the wind profile between 40 and 200 m with 5 m resolution. RAW data was recorded to enable additional analysis on turbulence and boundary layer height. The sodars were all validated at the 180 m tower or another 100 m tower at a nearby site before deployment.
One of the remote sensing points was equipped with a Leosphere V1 lidar (40-200 m), a Leosphere WLS70 lidar (100-1500 m) and a Vaisala ceilometer backscatter lidar, with the aim of covering the wind profile in the whole boundary layer, and using the ceilometer backscatter information to monitor the boundary layer height. The two Leosphere lidars suffered from different mechanical problems, mainly with the wiper and the cooling which caused low signal to noise ratio. The problems also caused low availability. In order to increase the availability, new statistics were calculated from the radial velocities form each pulse, fitting Gaussian distributions to the radial velocities and calculating the mean wind speed from the center of the Gaussian distributions.
The remote sensing campaign ran between April and October 2016, but some of the instruments were removed in August.
A figure showing the wind profile from the tower and the remote sensing profilers can be seen in Figure 6, and based on the overlap of the 95 % confidence intervals the instruments are considered to measure the same wind speed.

Pressure gradient measurements
An earlier Benchmark study over forests (Ivanell et al 2018) highlighted the problem for modelers of adjusting the pressure gradient or geostrophic wind in order to match a target wind speed at some height. In an attempt to remove that complication by measuring the static pressure gradient a system of four DigiQuartz 6000-16B-IS microbarometers was deployed in a square with around 3 km sides. To minimize the bluff-body and dynamic-pressure effects the sensors was placed in wind shielded locations beneath the canopy. Comparisons of the calculated geostrophic wind speed to the wind speed above the boundary layer as measured by the profilers showed that the pressure gradient system was able to measure the static pressure, but noise in the data require a sufficient averaging time to be used.

Measurement Campaign: 

Period, sampling, wind conditions range, etc.


The Hornamossen site was equipped with flow model validation in mind. Much of the motivation for the experiment design comes from an earlier experiment with a 140 tall tower over a similar forest, but with much less topographical complexity preceded this experiment. The main results from that experiment can be seen in Bergström et al 2013 and Arnqvist et al 2015. One conclusion was that the boundary layer height is a promising scaling factor for many of the characteristics of the wind important for wind energy i.e. wind turning, wind shear, and profiles of second and third order moments. That was the basis of recording raw data from the sodars as well as including the long range profiler and the ceilometer. Another conclusion from Arnqvist et al 2015 was that effects of stratification becomes very prominent on higher heights and was seen to effect almost all statistical moments as well as the roughness length which was shown to decrease in stable stratification. With that in mind, the benchmarks coming out of the Hornamossen experiment will focus on a variety of stratifications as well as transitions between different stratifications.


Complex terrain experiments in the New European Wind Atlas
J. Mann, N. Angelou, J. Arnqvist, D. Callies, E. Cantero, R. Chávez Arroyo, M. Courtney, J. Cuxart, E. Dellwik, J. Gottschall, S. Ivanell, P. Kühn, G. Lea, J. C. Matos, J. M. L. M. Palma, L. Pauscher, A. Peña, J. Sanz Rodrigo, S. Söderberg, N. Vasiljevic, C. Veiga Rodrigues
Phil. Trans. R. Soc. A 2017 375 20160101; DOI: 10.1098/rsta.2016.0101. Published 6 March

Wind power in forests: Winds and effects on loads
Bergström H, Alfredsson H, Arnqvist J, Carlén I, Fransson J, Dellwik E, Ganander
H, Mohr M, Segalini A, Söderberg S (2013) Wind Power in Forests. Tech. rep.,

Wind statistics from a forested landscape
Arnqvist, J., Segalini, A., Dellwik, E.and Bergström, H
Boundary-Layer Meteorol (2015) 156: 53.

Microscale model comparison (benchmark) at the moderate complex forested site Ryningsnäs
Ivanell, S. and Arnqvist, J. and Avila, M. and Cavar, D. and Chavez-Arroyo, R. A. and Olivares-Espinosa, H. and Peralta, C. and Adib, J. and Witha, B.
Wind Energy Science Discussions 2018 DOI: 10.5194/wes-2018-20