BOSTON — As SODAR (SOnic Detection And Ranging) and LIDAR (LIght Detection And Ranging) technologies mature, more developers are using them to support wind resource assessments. These remote sensing instruments are relatively easy to use and move from site to site. They provide hub-height wind speeds and direction, vertical wind speeds, and wind shear and veer above the heights of typical meteorological tower (met mast) measurements. They may also provide some measure of turbulence levels.
Unfortunately, users may fail to realise the full value of their remote sensing equipment due to a lack of clarity about best practices. For example, the portability of remote sensing equipment has encouraged many users to measure for short periods at multiple locations in order to capitalize on their investment. However, measurement periods that are too short, poor site conditions and a lack of verification of equipment performance can lead to data which are of limited or no use in an energy assessment.
Guidelines on the use of remote sensing measurements support wind resource and energy assessments. In this context, such assessments are evaluations of the resources available at a project site and are thus evaluations of the generation potential of wind turbines planned for that project.
Using remote sensing data in energy assessments
In an energy assessment, remote sensing data may be used to evaluate the accuracy of extrapolations from tower data; shear coefficients to be used with tower data; hub-height wind speeds and directions; and wind resource variability across the site.
How the data are used in an energy assessment will depend on the quality, quantity and representativeness of the validated data. The degree to which the data improve or reduce the uncertainty of the energy assessment will depend on, among other factors, the data recovery and its correlation to nearby met tower data, the data collection durations and the data collection locations relative to met tower and proposed turbine locations. Other project characteristics will also determine the impact of remote sensing data on an energy assessment. For example, if there is little uncertainty in the energy assessment without remote sensing measurements, then remote sensing measurements may have a marginal impact on the overall assessment due to the uncertainties inherent in the remote sensing data. Seasonal variability of the atmospheric conditions at the project may limit the impact of short-term remote sensing measurements on the energy estimate or its uncertainty.
Important aspects of remote sensing technologies which must be taken into account to ensure that data are useful for energy assessments include:
- Remote sensing technologies are still maturing. Configuration and software changes may affect accuracy, quality, or consistency.
- Measurement quality may be affected by site positioning. Even perfectly operating Sodars or Lidars may provide incorrect measurements in the presence of complex flow (spatially uneven flow) above the instrument. Complex flow may occur in complex terrain or near surface roughness transitions. Additionally, Sodar measurements may be affected by site-specific ambient noises and echoes from nearby objects and vegetation (ground clutter).
- Remote sensing measurements may be different than those of anemometry. Anemometers provide averages of point measurements of wind speed irrespective of wind direction (scalar averages). Conversely, Sodars and Lidars measure vertical, lateral and horizontal wind speeds. These are usually transformed to provide ‘vector averages’ of wind speed, although some instruments may provide scalar averages. In turbulent conditions, vector averages are lower than scalar averages. These differences mean that remote sensing and anemometry may not provide the same wind speed values, although each may be measuring correctly. Each type of measurement system may also provide slightly biased measurements under certain conditions. For example, in conditions with high shear the volume averages of remote sensing instruments may be lower than point measurements.
If these aspects of remote sensing technology are appropriately considered in a measurement campaign, remote sensing data can provide valuable information for an energy assessment. Compliance with site positioning, documentation and verification best practices provides the information required to evaluate the quality of the data and its value with respect to an energy assessment. The value of the data will depend on many factors but can be enhanced with careful planning and consultation. However, while guidelines cannot ensure how data will be used, attention to the following will ensure, as much as possible, that the value of data will not be jeopardised by a lack of awareness of the needs of the energy analysis process.
Requirement for on-site met tower
There should be on-site wind resource quality met tower data to confirm correct equipment operation. Tower data are used to identify outliers caused by operational problems and electrical or acoustic noise and to identify the effects of ground clutter, echoes, and complex flow. The tower need not be directly beside the remote sensing equipment, but should be within the footprint of the project and at a location with a similar wind resource.
DNV strongly discourages the use of stand-alone remote sensing data collection when used for future energy assessments. At least one nearby wind met tower should be used to provide confidence in the positioning of remote sensing instruments.
Avoiding the effects of nearby obstacles and noise
In forested areas, when measuring with a Sodar, trees must be cleared in a large circle around the instrument to avoid ground clutter contamination of data. Clearing trees to a distance 20 meters greater than the highest measurement height is the most conservative approach. Depending on the beam orientation and the instrument these conditions may be relaxed. In forested areas, when measuring with a Lidar, a clearing with a radius as great as the height of the surrounding trees is adequate.
Similarly, locating a Sodar 20 meters farther from solid structures that could reflect signals than the greatest desired measurement height is the most conservative positioning approach. Depending on the instrument design, these conditions may be relaxed. If possible, the Sodar should also be oriented such that the acoustic beams are not directed toward these objects.
Avoiding complex flow effects
Locations with complex flow (often caused by terrain or non-uniform roughness) should be avoided, if possible. These locations feature changes in flow conditions within the volume used to measure the wind, roughly a circular volume with a radius that may be between 20 and 60 meters and a thickness that may be 1-20 metres, depending on the instrument and measurement height. If the spatial variation of the horizontal or vertical wind speeds is expected to be greater than 1% over the measurement volume, a more detailed analysis of possible measurement biases should be conducted.
Among a range of requirements, the remote sensing documentation should include: instrument placement coordinates and elevation to within a few meters, including datum and confirmation that the instrument is level (within ± 0.5°) as well as beam orientation with respect to true north; instrument configuration, manufacturer, model, and serial number, measurement cone angle and confirmation that the clocks of the anemometer and remote sensing loggers are within five seconds of each other. If a Sodar is used, confirmation, if possible, that the transmit frequency is correct, that any temperature sensor is operating correctly and that any software or firmware is using the measured ambient temperature for calculations.
Users may fail to realise the full value of their remote sensing equipment (Source: Seewind)
Site condition document requirements include: photo documentation of surrounding terrain and surface features in 45° direction sectors; azimuth, distance, height, elevation angle and subtended angle, as seen from the remote sensing instrument, of surface features and obstacles to flow within visible range; azimuth, distance and elevation angle, as seen from the remote sensing instrument, of the nearby tower. If a Sodar is used, documentation of possible sources of echoes and ambient noise sources, such as nearby roads, crickets, etc.
Sensor locations and configuration documentation should cover the tower location and elevation to within a few metres, the tower height and dimensions, instrument manufacturers, serial numbers and calibration documentation, as well as sensor installation heights, boom dimensions and azimuth angles. Finally, tower site documentation should include photos of the surrounding terrain and surface features.
Verification refers to comparisons against co-located tower data that are used to verify the correct operation of the remote sensing instrument. The accuracy of comparisons with the tower data is dependent on the quality of the tower sensors and is limited by spatial variation of the wind between the tower and the remote sensing device should they not be co-located. Verification against accurate sensors and at sites in uniform terrain and surface roughness conditions will enable analysts to better evaluate the quality of the data and may reduce uncertainties. This could be of critical importance and a verification test every year or within six months of any data collection, or after any software, hardware or firmware change is recommended.
Planning and implementing remote sensing
Energy assessments are usually based on one or more years of met tower data from one or more met towers within a project area. These data are then used to estimate the wind resource at turbine locations and hub heights. There may be enough topographic variability of the wind resource that the estimates of wind speeds at some turbine locations are quite uncertain. There may also be uncertainty as to whether the shear exponent as measured using tower data represents the shear behaviour above the tower or at the proposed locations.
Remote sensing data can be used to define the relationships between the longer-term tower data and the wind resource at remote sensing measurement sites and heights. A well-planned measurement campaign addresses the sources of uncertainty that are expected to be the greatest and thus provides a data set that is as representative as possible of the annual conditions at the site.
Often a remote sensing instrument is deployed at one or more sites for less than a year. The instrument might be returned to a previous measurement site one or more times in a year to ensure that annual conditions have been measured (this is referred to as seasonal sampling). If seasonal sampling is not employed and if data are collected only for part of a year, there may still be uncertainty about how well these results represent behaviour over a full year. This may be acceptable if the data, in spite of the uncertainty, provides valuable new information.
An examination of numerous data sets indicates that seasonal variations may be significant at many sites. To adequately characterise the conditions over a year, either seasonal sampling or longer measurement durations are usually needed. Short-term results may or may not characterise the annual average wind resource at the measurement site, depending on the nature of seasonal changes in wind conditions. Even though measurements may characterise the short-term measurement period quite accurately, they may have an uncertainty of up to three percent when compared with annual results. Reducing that uncertainty requires measuring over multiple seasons. Two methods for reducing seasonal uncertainty are recommended: measuring for at least eight months or more and seasonal sampling with short-term deployments. Year-long measurements should be considered as they provide a clear picture of shear across the rotor and hub-height wind speeds under all expected conditions.
Planning instrument deployment
The sequence of measurement locations and approximate measurement durations should be determined in advance. Available information on monthly or seasonal variations in wind speeds and directions, shear and ground cover should be used to guide the timing of measurements at each location. Seasonal sampling is important to include in a measurement plan if there are reasons to believe that the relationship between the wind resource and the tower varies significantly over the year. Seasonal changes could be due to changes in surface roughness, ground cover, stability, foliage, wind direction or solar insolation. In spite of the apparent advantages of seasonal sampling, careful consideration should be given to planning for a short initial data collection period on the assumption that later data collection will provide the rest of a useful data set. Possible operational problems, weather-related access or delays getting crews to the site should be taken into consideration.
The need for standards
Guidelines for positioning at site, documentation and verification of remote sensing instruments and for planning and implementing remote sensing measurement campaigns help to ensure that the results have the greatest value in an energy assessment.