The Town of Newbury may soon be joining the growing ranks of communities that are building wind turbines, and hopefully it will take to heart the lessons that are quickly being learned by its neighbors.

Newbury is looking at three sites — the former town dump, anywhere along the Great Marsh, and a 5-acre plot of land on Plum Island. It has not yet been decided which, if any, of these sites will be chosen.

There's a general agreement that America needs to find better ways to create electricity, and wind power is certainly one of the viable methods. But wind is a fickle thing, and it's becoming clear that the practice of building wind turbines isn't always in sync with the reality of the wind's habits.

Case in point is the trio of small wind turbines in Amesbury at Cider Hill Farm. A report from 2008 showed that one was producing 29 percent of expected power, the other at 22 percent. The third turbine had not been completed at the time of the report.

In Newbury, a 10-kilowatt turbine off Route 1, built and installed with the help of a grant from the Massachusetts Technology Collaborative, produced only about 25 percent of its anticipated output.

The disappointing performance of small wind turbines has caught the attention of Massachusetts Technology Collaborative, which provides grants for many of these projects. The collaborative estimates these "small" turbines are producing only about 30 percent of the power that was originally projected. There are a number of reasons why they have not performed well, but one of the most common reasons is placement. There's simply not enough wind to provide the desired results.

Larger turbines, like the 292-foot-tall Mark Richey turbine in Newburyport, tend to be more efficient. But proper placement for large turbines is vitally important. There is no recent power generation data posted on the Richey turbine, but state wind data indicates it is not located in the area's strongest wind zone.

Local wind turbine zoning laws often look only at the ground impact of wind turbines, such as their proximity to homes and businesses. It's an important aspect, of course, but the predominant measure should be how much wind the area can expect. And by that measure, the Technology Collaborative's data makes it crystal clear that there is only one area in this entire region where wind turbines make sense. That area is our coastline —- specifically Plum Island, Salisbury Beach and the easternmost edges of the Great Marsh, the enormous marsh that separates the beaches from the mainland (the maps are available online at the collaborative's Web site).

Newbury is considering this area, which makes sense. But our local wind turbine zoning laws should take it a step further. We should restrict the construction of turbines to areas where wind data shows they will be effective, then apply the rigorous ground zoning — such as minimal setback from homes — and see where turbines would be acceptable.

Newburyport has learned many lessons from the neighborhood nuisances caused by the Mark Richey turbine. City officials are redrawing the zoning laws to take care of those problems, but perhaps they should also take a look at the turbine's power output to see if it makes any sense at all to have another turbine in the industrial park. Data collected by Rochester Electronics, which considered building one, showed it's not viable.

If wind turbines are to be built, we should build them in places where they make sense for people's lives and the wind's habits.