Numerical simulation of wind power potential in Upstate New York

Meteorologists at SUNY Oswego have been running the Weather Research and

Forecasting (WRF) model every day to produce high-resolution numerical

simulations of meteorological conditions over much of upstate New York. The

purpose of these simulations is to identify the potential for wind energy in

the region and to create a 'climatology' of wind speed, wind direction and

temperature at 10, 40 and 80 meters above the ground. We began archiving hourly

output from WRF in February 2009 and we will continue for at least the next two

years to create our data base. WRF is running on a doubly-nested grid to

ensure that both large-scale meteorological forcing and local geographical

effects are well-represented. The large domain has 12-km grid spacing covering

an area from eastern Iowa to western Labrador and from northern South Carolina

to James Bay. The intermediate grid has 4-km resolution and covers the eastern

Great Lakes region, mid-Atlantic states and most of New England. The fine grid

has 1.333-km resolution and covers an area 320 km by 200 km centered at Oswego.

The model has 33 vertical levels in a terrain-following system where the lowest

sigma levels correspond to 10m, 40m and 80m above ground under typical

meteorological conditions. Initial data and boundary values (updated every

three hours out to 24 hours) for the large domain are obtained each evening

from the 0000 UTC run of the operational North American Mesoscale (NAM) model

available online from the National Centers for Environmental Prediction (NCEP).

During Summer 2009, a meteorology major, funded by a grant from the

Department of Energy, is archiving WRF output and comparing WRF-simulated wind

and temperature predictions with observations from seven regularly-reporting

surface stations in the Lake Ontario region including Oswego. We expect to use

our recently acquired mobile tethersonde system later this year to measure

winds at five levels in the lowest 150 meters above ground to help validate WRF

predictions of wind and temperature profiles for a variety of atmospheric

conditions. WRF-simulated soundings will be compared with the soundings

obtained near Oswego using our Vaisala upper-air tracking system.

The WRF predictions of 80-m wind speed are being used to compute hourly and

monthly averages of theoretical wind energy generation for hypothetical 1500 KW

wind turbines located at all of the 36000 grid points on the fine grid. Similar

estimates of wind energy generation from small (5 KW) turbines are being made

using the 40-m WRF winds. These computations will be used to identify sites

which have the greatest wind power potential. Our preliminary results for the

period February-July 2009 show that theoretical power generation ranges from 25

to 45% of rated output with the greatest values occurring over the interior of

Lake Ontario. This does not include losses associated with turbine design. Over

land the greatest values appear to be found in regions of highest terrain

(e.g., the Tug Hill Plateau and the hills south of the New York State Thruway).

Wind direction may play a significant role. Plots showing wind vectors and

simulated wind power contours for a few selected days indicate that wind power

potential is greatest in the areas of steepest downslope.