by Maggie Kerchner, Air Resources Laboratory
The ceilometer, in the foreground, is surrounded by a barrier to keep local livestock from knocking it over.
Do you remember learning in science class that the atmosphere is divided into four layers: Troposphere, Stratosphere, Mesosphere, and Thermosphere? Perhaps you recall learning that all the Earth’s weather occurs in the Troposphere. But did you know that the lowest portion of the Troposphere, called the Atmospheric Boundary Layer (ABL), is the area most in contact with people and its behavior is directly influenced by what exists and occurs at the surface? Topographical features (e.g., water bodies, mountains, trees, and buildings), surface temperatures, large-scale weather patterns, and cloud cover all play a role in determining the behavior of the ABL. Yet, it’s the most difficult part of the atmosphere to simulate with models. This is because there are many complex interactions occurring between the Earth’s surface and the lower atmosphere. Improving our scientific understanding of the ABL can help our society in many ways, such as improving wind forecasts for the growing renewable wind energy industry.
Texas, the state that produces the most wind energy in the U.S., is where NOAA Air Resources Laboratory (ARL) scientists have been applying their expertise to gather low level wind and turbulence data as part of cooperative research with Duke Energy Generation. The Ocotillo Wind Farm in west Texas stretches across 2500 acres and produces a total of 58.8 megawatts of power from 28 wind turbines. According to Will Pendergrass, with ARL’s Atmospheric Turbulence and Diffusion Division (ATDD), the project has been a perfect opportunity for ARL to advance its atmospheric boundary layer science and at the same time help improve wind power forecasts. The agreement allows ARL access to Duke Energy’s Ocotillo Wind Farm to collect much-needed measurements. The data are then made publically available on ARL’s website for Duke Energy Generation and other wind energy industries to use. The data are also available to the broader atmospheric research community. Pendergrass explains that this is the first time an operating wind farm has direct measurements at the height of the turbine blade to allow for an understanding of the model uncertainties with NOAA’s operational wind forecasts. For the industry, knowing more about the uncertainties translates into confidence in the models and this confidence allows them to better manage the power production system.
A radiosonde launch on the wind farm to observe how atmospheric flow changes in both space and time.
Unlike natural gas or coal, wind is an intermittent source of energy, and wind farms can’t produce a steady supply of power. If they produce too little wind power, they can’t make money. If they produce more then what can be used, the industry risks losing money because the excess energy cannot be stored profitably yet.
The challenge for power utilities and grid operators is to predict as accurately as possible how much wind energy they can expect to produce throughout the day or over the course of weeks or even months. The better the forecasts of wind power production (or the fewer uncertainties), the lower the cost to power producers and to rate payers. In order to have better wind forecasts, high quality ABL data must be collected and then used by wind forecast models.
A Full Suite of Measurement Approaches
A SODAR system positioned near the wind turbines on the Ocotillo Wind Farm.
On the Ocotillo Wind Farm, ATDD scientists have been collecting measurements of wind speed, wind direction, and atmospheric turbulence from sensors placed at varying heights on their 30 meter high research tower next to an 80 meter high wind turbine. Duke Energy researchers also placed sensors at varying heights on a 80 meter high tower, next to the same turbine. Combining all of the data from the various heights shows how the wind is behaving from the ground to the height of the wind turbine blades. According to Will Pendergrass, there are few researchers who collect wind profile data like this. ARL’s approach is unique because data are collected over long periods of time and at different heights on the tower.
While the sensors on the towers have been busy collecting data near the ground, the ATDD scientists employed a measurement technique that put them higher into the air. A SODAR (SOnic Detection And Ranging) system was positioned near the research towers for about three months. SODAR systems are used to measure the turbulence and winds of the lower layer of the atmosphere by measuring the scattering of sound by atmospheric turbulence. SODAR data can be collected a few hundred to several hundred meters high, but SODAR can be quite loud. Therefore, operation is usually short-term. ARL’s SODAR system was able to acquire data at varying heights from 50 meters to about 250 meters.
A second, short-term measurement technique was then applied to give the research team insight into how the atmospheric flow around the area of the turbine changes in both space and time. For this, ATDD scientists launched radiosondes (or weather balloons). The radiosondes carry a small, lightweight instrument package that transmits information on wind speed, wind direction, temperature, and humidity, in real time back to a ground station. Radiosondes, which can travel up to 400 meters high, are particularly useful for gathering data on the low-level jet, a fast moving ribbon of air in the lower levels of the atmosphere. The jet generally forms during the evening, strengthens during the course of the night, and dissipates shortly after sunrise as the air begins to mix from the warming of the surface. Low level jets are a phenomenon common to the Great Plains.
From left to right, the Duke tower, the wind turbine, and ARL’s research tower.
In the coming months, ARL will begin to apply another, short-term measurement technique — the use of a ceilometer. A ceilometer is a light detection and ranging laser (LIDAR) system traditionally used to determine the height of a cloud base that can also be used to monitor ABL structures. Ceilometers work by shining a laser into the atmosphere.
In ABL science, scientists must use a variety of techniques to fill in different pieces of the puzzle. With improved resolution of wind forecast models, it has become more critical that the models include data, such as that coming from ARL’s full suite of measurement techniques and instruments. Only by using these kinds of data can wind forecast models be truly tested and evaluated. For the first time at NOAA, ARL has provided information to the wind power industry which addresses the uncertainty in NOAA’s forecast models. Reduction of this uncertainty is a critical element which will be used by the industry to improve wind power generation efficiency.