The Askervein Hill project was a collaborative study ofboundary-layer Bow over low hills carried out under the auspices of the International Energy Agency Programme of R & D on Wind Energy Conversion Systems. Two field experiments were conducted during September-October 1982 and 1983 on and around Askervein, a 116 m high hill on the west coast of the island of South Uist in the Outer Hebrides of Scotland. During the experiments, over 50 towers were deployed and instrumented for wind measurements. The majority were simple 10 m posts bearing cup anemometers but, in the 1983 study, two 50 m towers, a 30 m tower, a 16 m tower, and thirteen 10 m towers were instrumented for 3-component turbulence measurement.The present paper provides an overview of the project as a whole, including details of the instrumentation and a summary of the data obtained. Additional papers in the series, which are to appear in this journal*, will consider different aspects of the experimental data and related numerical-model and wind-tunnel studies.

Eulerlan time-delay correlation tensor (Sec. 2.1) t time T temperature T. Integral time scale of turbulence (Sec. 2.2) U(t) total velocity component In x-dlrectlon 2 Ö + u(t) Ü mean velocity In x-dlrection U(t) turbulence velocity vector * i u(t) + Jv(t) + k w(t) Ü friction velocity Ö_ gradient wind velocity (Sec. 3.1) u(t), v(t), w(t) fluctuating velocity components In x, y and z directions, respectively u', v', w' rms values of fluctuating velocity components; u* s^u ,v ,s «/v , w'svw x, y, z Cartesian coordinate axes z height above earth's surface z roughness length (Sec. 3«^)

A detailed analysis has been carried out of the temporal and spatial structure of mean winds and turbulence in the neutrally-stable planetary boundary layer over typically rural terrain. The data were obtained from a horizontal array of tower-mounted propeller anemometers (z = 11 m) during a five-hour period for which the mean wind direction was virtually perpendicular to the main span of the array. Various turbulence characteristics have been obtained for all three components of velocity and have been compared with idealized models for such a flow and with some of the other available atmospheric results.Considerable tower-to-tower and block-to-block variability has been observed in many of the measured results, particularly in those for the horizontal-component integral scales. Surface shear stress, roughness length and turbulence intensities were in good agreement with expected values for such a site. Power spectra for all components displayed significantly more energy at middle and lower frequencies than that observed by Kaimal et al. (1972) over flat, relatively featureless terrain. This is felt to be a result of the generally rougher gross features of the terrain in the present case and has led to the development of a modified version of the Kaimalspectral model which fits the observed data better than either the original Kaimal model or the von K&m&n model. It is suggested that it may in future be possible to represent power spectra over a wide range of terrain types by using such a modified spectral model.Integral scales of turbulence were calculated by three different techniques and in most cases displayed a strong dependence on the technique used. Averaged values of scale showed reasonable agreement with most of the available atmospheric data and with the values suggested by ESDU (1975). The anticipated elongation of turbulent eddies in the longitudinal direction was confirmed for all three velocity components, although it was found to be not as large as some other observations. Nomenclature ai, h k '4 KlO I:' n n, Fig Ri CAY, 7) {i (7) &j(O) Si (n) spectral equation constants -see Section 3.5 non-dimensional frequency = nz/ fi wave number = n/ 0 wave number at which peak of i-component power spectrum occurs surface drag coefficient (z = 10 m) = (u*/ ii,,)' integral scale of i-component for separation in j-direction, calculated by correlation integral technique (Section 3.7) integral scale of i-component for separation in j-direction, calculated by exponential fit technique (Section 3.7) integral scale of i-component for separation in j-direction, calculated by spectra1 fit technique (Section 3.7) frequency filter cut-off frequency (-3 dB) gradient Richardson number normalized time-delayed cross-correlation of i-components for separations in lateral direction [e.g. d, (Ay, 7) = ~/(cr,,~,)] normalized autocorrelation of i-component [e.g. g,,(7) = u(t) u(t + 7)/m:] Reynolds stress coefficient for i-and j-components [e.g. d,,(O) =G/((T,(~,)] power spectral density of i-component time Boundary-Layer Meteoro...

This is one of a series of papers on the Askervein Hill Project. It presents results from the Askervein 1982 and 1983 experiments in the form of vertical profiles of mean wind and turbulence integral statistics at upwind reference locations and at two hilltop sites. The data were obtained from a variety of sensors including sonic, Gill UVW and cup anemometers mounted on 50,30,17 and 10 m towers and TALA kite systems. Comparisons with numerical-model predictions are discussed.

Wind-tunnel simulations ofneutrally-stable atmospheric boundary-layer flow over an isolated, low hill (Askervein) have been carried out at three different length scales in two wind-tunnel facilities. The objectives ofthese simulations were to assess the reliability with which changes in mean wind and turbulence structure induced by the prototype hill on boundary-layer flow can be reproduced in the wind tunnel, and to determine the relative impact of certain modellmg approaches (surface roughness, model scale, measurement techniques, etc.) on the quality of the simulations. The wind-tunnel results are compared with each other and with full-scale data and are shown in general to model the prototype flow very well. The effects of relaxing the criterion of aerodynamic roughness of the model surface were limited to certain regions in the lee of the hill and were linked to separation phenomena.