NB: this article is part of Renewable Energy Focus’ Science & Technology series (subscribe to the magazine here for access to articles as and when they are published).

The GIS module's purpose is to locate potential sites for wind and solar farms based on geographic and regulatory characteristics, and to characterise the potential for biomass use. This information could be used to analyse renewable energy policies and investments, according to the researchers writing for the journal Renewable Energy (click here for the full unedited paper).

The study focuses on large-scale, utility-grade solar and wind installations due to the efficiencies that can be achieved at larger scales, but also because small-scale and domestic installations often require very localised policy considerations. For biomass, the study looks at solid wood waste in the context of co-firing with coal.

Based on 2007 US Energy Information Administration (EIA) figures, the Appalachians have 148 electricity generating facilities producing a total of 198,474,165 MWh in baseline annually. The region was chosen for the GIS study because it is dominated by coal-fired energy generation, while at the same time it contains some of the best onshore wind power potential in the Eastern U.S.

Currently, renewable energy generation consists of hydropower, biomass, wind and landfill gas – totalling 3.15% of total generation, plus an additional 0.22% of co-fired generation - estimated to be derived from biomass.

In terms of incentives, only North Carolina has established a binding renewable portfolio standard (RPS), whereas Virginia has a voluntary RPS. There are already a small number of wind and solar farms proposed for future use, but locations have not yet been approved and were therefore not included in the study.

Generation by Source Within the Region

Biomass co-firing potential

The study focuses on solid wood waste, including urban wood waste, primary and secondary mill residue, and forest residue. These allow for the cheapest and easiest utilisation within the region. Such biomass can also be co-fired with coal, causing minimal disruption to the generation process - and also requiring minimal investment.

A total of 31 coal plants that could be used for co-firing were identified using the GIS model. Furthermore, a National Renewable Energy Laboratory (NREL) dataset specified an estimated 9,571,545 tonnes of solid wood waste within the region.

If used for co-firing in coal plants, biomass could replace 8.7% of coal use in the region. Furthermore, the model estimates the retrofit of coal plants for biomass co-fire to cost US$100/kW installed. Based on the installed capacity of almost 19 GW at the 31 coal plants in the region, and if 8.7% of that capacity was dedicated to biomass co-fire, rough estimates suggest it would represent a capital investment cost of US$164,961,100.

Solar PV could reach 5 GW

The amount of solar insolation within the southern Appalachians region is in the range of 4.19-5.01 kWh/m2/day annually.

Areas with a slope of under 2.5% are acceptable because the relative flat surface means a southern exposure is not required as the solar panels can be tilted. Another good combination would be areas with slopes of 2.5%–15%, with south-facing aspects (112.5°–247.5°). A location with a slope greater than 15%, regardless of exposure, is seen as undesirable for solar farm development.

Another criteria in the siting of solar farms is current land use, as one criticism of solar farms are that they take up large areas of land that could be put to other uses – often uses that would be more economically beneficial to landowners. Whereas wind farms can be sited on agricultural land without having a big impact on the amount of productive land, solar farms require a majority of the land to be used for solar only.

Due to this, the study looks at what the National Land Cover Dataset (NLCD) describes as “barren land” – i.e. areas of bedrock, desert pavement, scarps, talus, slides, volcanic material, glacial debris, sand dunes, trip mines, gravel pits and other accumulations of earthen material. Vegetation would typically account for less than 15% of total cover. However, this limits the potential land for solar farms to 0.41% of the region.

The final criterion is conservation concerns - as classified by the Protected Areas Database of the U.S. Combining these criteria with the size requirements (here, only locations over 10 acres qualify), means the total number of potential locations is limited to 477.

The maximum installed solar capacity using the model would be just over 5 GW, representing almost 6.6 TWh of annual production. The overall generation from solar could potentially replace 3.33% of the total baseline generation. Installing the full capacity would, however, require an estimated total capital investment of over US$32.6bn and an average generation cost of US$0.2219/kWh.

Wind restrictions

For wind farms, the study used NREL's criterion of a having a wind power class three or greater to be considered acceptable for wind farm development. By this criterion only, the potential land in the region has been restricted to 2.38%.

Percentage of land within the region by NREL wind power class

In addition comes the question of sloping – although not as strict as for solar developments – a slope above 20% would cause difficulties for installation and the stability of the wind turbines. The NLCD classifications that are permissible in the wind farm model include barren land; the forest types scrub/shrub, grassland, pasture/hay; and crop land. However as wind can be co-located with agricultural activities, the number of NLCD classifications permittable for wind farms is greater than in the solar scenario.

One of the three forest classifications make up 68.82% of the region and 92.5% of the land with wind power class three is within this subset.

Wind is also affected by conservation restrictions, but unlike solar, wind has further constraints due to visibility and potential noise. Wind turbines can also have an impact on urban land use and airports. This requires so-called buffer zones between a wind farm and the location of a specific land use or conservation area. Buffer distances are subject to local regulations, which could not be included in the study.

By using NREL's buffer conditions, 99.995% of all land within the region would be eliminated from consideration. When adding the constraints on slope and wind power class, no land is left for wind farms. The study therefore uses a small buffer for “Developed Open Space” land versus “Developed, Low/Medium/High Intensity” land, which reflects the fact that many of the land uses classified in the former category are roads, which would not impede wind farm development.

The one operational wind farm within the region is in fact on land that is constrained by the buffers in the GIS model, and other planned wind farms do not meet the constraints present in the model. The study therefore recommends that more detailed planning is carried out at local levels.

By taking all criteria into consideration, the model found 28,360 possible good wind farm locations. Each area is 120 acres or more. A value of 40 acres per wind turbine is seen as the minimum necessary for the GIS model.

Based on the constraints in the GIS model, there is an estimated capacity of over 1.8 TWp that could be installed in the region, capable of an annual generation of over 6.4 TWh. This would represent 3.24% of the baseline amount. However installing the full capacity of wind power would require an estimated total investment of around US$13.4bn and an average generation cost of US$0.1466/kWh.

Of the 202 potential wind sites, 46 have an estimated generation cost under US$0.06993/kWh, whereas the biggest cost for a potential site is US$0.4691/kWh.

Most of the potential wind sites are located in the upper western portion of Virginia and the bordering areas of West Virginia. There are no potential sites in Kentucky, North Carolina or Tennessee. In Kentucky and Tennessee, this is mainly due to a lack of sufficient wind speeds.

Conclusions

If the full potential of the renewable energy sources identified in the GIS model are realised, the total capital investment required would be US$46.16bn, replacing 15.27% of baseline generation. Combined with the current 3.25% of renewable energy generation, the total renewable energy generation would reach 18.52%.

However, as no ‘one’ renewable energy came out as a substantial, long-term solution to increase the share of renewables, the study recommends exploring multiple sources of renewable energy to meet emissions reductions and energy independency targets.

Estimated Cost per kWh by Source

NB: This article was published in Renewable Energy Vol 36 (2011), by Andrew N. Arnette and Christopher W. Zoebel, Spatial Analysis of Renewable Energy Potential in the Greater Southern Appalachian Mountains, Pages 2785-2798, Copyright Elsevier (2011).

Renewable Energy, official journal of the World Renewable Energy Network (WREN), seeks to promote and disseminate knowledge of the various topics and technologies of renewable energy. It is aimed at assisting researchers, economists, manufacturers, world agencies and societies to keep abreast of new developments in their specialist fields and to unite in finding alternative energy solutions to current issues such as the greenhouse effect and the depletion of the ozone layer.

FULL ARTICLE: Spatial Analysis of Renewable Energy Potential in the Greater Southern Appalachian Mountains

About:

Kari Williamson is a freelance journalist specializing in renewable energy. She is the former Assistant Editor at Renewable Energy Focus and Renewable Energy Focus U.S.

Renewable Energy Focus, July/August 2011.