Five years ago, First Wind’s in-house meteorological team zeroed in on a couple of new potential wind farm areas in Maine that were different: moderate-elevation plateaus instead of the traditionally preferred high-elevation ridgetop locations. At the time, those sites looked like “a little bit of a gamble,” recalled Dave Fowler, First Wind’s director of development for the New England region. But the company put up a few meteorological (MET) test towers at those sites, and after a couple of years added newer remote-sensing technologies to broaden the wind data collection – and found there was “really impressive wind resource” at those locations, added Matt Kearns, First Wind’s VP of development for the Northeast region. One of those sites became the company’s 34-MW Bull Hill operation, which came online in the fall of 2012. The other, officially pitched to Maine state officials this summer, is the proposed 186-MW Bingham Wind project in southern Aroostook County, which will be the largest wind farm in New England upon completion.
Without the data provided by newer remote-sensing technologies, First Wind likely wouldn’t have pursued projects on these new lower-height locations, which are opening up “whole new areas in Maine” for potential development, according to Fowler.
MET towers and anemometers have been the workhorse wind assessment technology for decades, gathering wind data at relatively minor expense. The MET tower approach is still highly reliable and most stakeholders will accept a well-executed MET tower-based campaign, according to Matthew Filippelli, lead engineer at AWS Truepower. Now, though, remote sensing technologies such as sodar (SOnic Detection And Ranging) and lidar (laser light plus radar) are expanding the scope of the wind assessment data, and thus the confidence in a site’s wind energy resources.
Much of the increased adoption of remote sensing is driven by the industry’s shift toward bigger wind turbines, with taller hubs and longer rotors. By the end of this decade, a third of wind turbine installations in Europe will be IEC Class III lower-wind-speed turbines in the 80-100 meter range, calculates Feng Zhao, managing consultant with Navigant Research. In the U.S., developers including First Wind are moving into regions like New England that have untapped areas with lower wind regimes, and deploying taller turbines to tap the higher and more reliable winds that make these sites feasible.
At only 60-80 meters in height, traditional MET towers require extrapolation to calculate wind resource data up at these new wind turbine heights. More extrapolation means less certainty about the site’s potential production, and uncertainty means increased risk and difficulty getting a project financed. Now, all wind evaluation methods are being applied to bankable energy investments, added Katy Briggs, head of DNV Kema’s energy analysis section: “Some come with measuring campaigns and accuracy of data, some maybe have more uncertainty in measuring wind speed data, but the industry is accepting the data.”
Comparing MET Options
As developers become savvier about wind resource mapping beyond standard MET towers, here are their choices:
Taller MET towers have the advantage of being based on the longstanding reliable low-cost technology, but they quickly lose that edge, with a decked-out 100-meter MET tower costing up to 10 times more than the standard MET tower, Zhao points out. Extra-tall MET towers also run the risk of running into permitting and approval problems from local authorities to the FAA. However, taller MET towers are built for a longer lifetime of on-site data collection, so they can be cost-effective for a long-term wind resource monitoring strategy, Briggs noted.
Tinkered with since the 1980s to characterize wake effects behind wind turbines, sodar gradually has become part of up-front wind site assessment campaigns. Relatively small, low-power (usually provided by a solar PV panel), and portable, these units can be quickly installed for early site assessment – even before deploying a MET tower – and moved around to expand data collection and the wind resource profile. Sodar devices also are “in a similar cost range” as MET towers, according to Briggs. Sodar isn’t as accurate or as scalable as lidar, though, with capabilities generally deteriorating above 120 meters, Feng points out. And its data capture and delivery can be eroded by various environmental conditions: obstructions (trees, buildings, steep hillsides), precipitation, and even high wind speeds that generate their own acoustics.
Lidar was developed in the early 1960s by combining lasers and radar to generate much higher-quality data between long distances. Its initial application was mapping the moon’s surface for exploration, then quickly adopted in aviation and meteorology, though wind energy assessment wasn’t investigated until the early 2000s. Lidar generally provides data with better accuracy, especially at greater heights (up to 200 meters), and is less susceptible to adverse site conditions. Advanced lidar devices can obtain data from several kilometers away, vertically or horizontally. However, lidar is significantly more expensive than sodar and traditional MET towers. It also tends to require more power (exceeding 100 W), which is problematic for sites that aren’t grid-connected or in case of an outage; these units generally require backup which adds even more cost.
Both sodar and lidar have faster setup and mobility vs. fixed MET towers, but this needs to be used in moderation. Industry consensus is that a one-year duration of measurements taken in a fixed location provides the best reduction of uncertainty, according to Robert Poore, senior advisor for renewable energy services at DNV KEMA. That assumes synchronous use with multiple onsite MET towers providing longer-term (2-4 years) data, providing the fullest picture of a site’s wind resource.
All MET options have some degree of susceptibility to environmental conditions, so the choice between them is very site-dependent. Sodar has trouble with nearby tall structures or objects that can interfere with sound signals. Lidar, being based on light, can’t penetrate fog. Cup anemometers on towers are susceptible to icing.
Operations and Beyond
The decision about what MET technology to deploy doesn’t end with site assessment and selection; it’s not as well-known that the technology has value throughout a wind farm’s lifespan for power forecasting and modeling, Filippelli points out. “Sodar and lidar get attention at construction, but MET towers provide a good, useful data source throughout the entire project lifespan.”
Remote sensing technologies, though, are getting “a lot more experience in operation now,” Briggs offered. “The kinks are getting worked out.” Side-scanning lidar, for example, likely will gain favor in forecasting at operational wind plants, for early identification and preparation of changes in approaching winds that could change power output.
And investors are getting on board as well. Poore indicated that there are examples where some smaller projects, or portions of larger ones, have been partially financed based solely on remote sensing measurements. “The time is coming,” he predicted, when entire larger utility-scale projects will pursue and obtain a financing decision based solely on remote sensing measurement technology. In most cases, though, he admits that “there will always be a MET tower involved in a wind project.”