Countries willing to deploy renewable energy technologies face a number of issues. Investments must be planned in advance, which requires an understanding of the share of energy mix that can be supplied by renewable energy sources; the energy costs;thetechnologiesthat are most adapted to their local conditions;and an assessment of the investments volume required, in terms of support schemes, human capacities, or related infrastructures. Investors prefer to have long-term commitments on market volumes and long-term perspectives in creating a supply chain.

These questions can be answered by precise information on the value and location of renewable energy potentials. The accuracy of the estimate of the potentials directly translates into a risk in the decision-making process, and also a risk in the national strategy decision process. In this, the accuracy of a renewable energy potentials estimate is one of the strategic elements in deploying renewable energy technologies.

Estimating the potentials requires large upfront investments in measurement campaigns, extensive consultations, and a high level of technical knowledge to assess the resource, to provide "bankable" information to support investments

The information on the resource is not sufficient for the decision-making process, which requires an evaluation of the technical and economic potential; and these account for the geospatial, engineering, economic and social constraints for renewable energy developments.

Are there standards for assessing renewable energy potentials?

There is a need to take a detailed look at the definition of renewable energy potentials, and the methods used to obtain realistic estimates. Within the literature, there is a high discrepancy in the approaches and outcomes in the evaluation of potentials. No detailed standard method exists for estimating the theoretical, technical, and economic potentials. The value of potential at a given location will depend on the quality of the resource data used as the input, as well as the methodology used to estimate the final value of the potential.

The literature provides the best estimates possible, considering the datasets available, and using a range of dimensioning parameters. However, different approaches, using different input datasets leads to significantly different results that are hardly comparable.

A major consequence of these differences is that the decisionmakers are faced with a number of assessments that used different methods and provide different results, and they request clarity on the key factors that influence their decisions.

In addition, although current measurement devices and models for estimating the resource with a high precision are continuously improving, the methods to estimate the potentials are not standardised, leading to significant variations in the final estimate. These discrepancies in approaches may negate the efforts put in improving the measurement chain: the efforts made towards improving the measurements and models for estimating the resource will be compromised if the methods in approaches for calculating potential are not standardised and improved as well.

Generally speaking, coarse datasets of wind and solar resource are used for global analyses and detailed datasets are used for local analyses. In the future, this trend might change, since modern computing capabilities enable the generation of detailed resource datasets with a large, and indeed global, geographic coverage. This allows for analyses over a large area to be made using approaches developed for local assessments, except that these detailed assessments require a large amount of ancillary data, which might not be available at large scale. A question arises on the impact and value of considering detailed information to improve the final estimate, and the value of investing in developing the missing datasets. Answering this point requires an understanding of the major driving elements for assessing renewable energy potentials, the sensitivity of the final results to one or more additional layers of information, and the sensitivity of the decision-making process to uncertainties in the estimate.

Download Pdf

The private sector assesses investment opportunities in terms of two interlinked elements: risk and uncertainty. Any actions that decrease either risk or uncertainty will decrease the cost of financing, decrease the cost of energy and ultimately make for a more desirable investment environment.

There are four categories of risk associated with a wind project: technology risk, credit risk, interconnection risk and production risk. The last has proven to be a particular challenge for the industry, resulting in significant historical discrepancies between actual and predicted yield. This situation has arisen for several reasons: developers have traditionally overestimated predicted energy yields, off-takers and financiers have not always conducted proper due diligence and modellers have had difficulty in estimating yields on complex terrain. However over the past three to four years the industry and its financiers have improved in addressing these issues and as a result the industry has seen a steady increase in the reliability of predicted yields.

The path to decreased risk starts at the prospecting stage where developers look to mesoscale models. With a high-quality map - preferably correlated with long-term data - a developer can quickly focus on areas of good potential, setting the stage for a detailed micro-siting exercise using calibrated met masts. If the set-up, monitoring and data analysis follow best practice, the result is a bankable yield estimate that provides comfort to lenders and lowers the cost of financing. Measurement standards and certification bodies only can guarantee the data are complete and accurate.

By providing high-quality mesoscale maps at a regional or national level, the relevant government or utility sends a clear signal that they are serious about providing a level playing field that reduces risk and uncertainty for project developers, ultimately providing for a better investment environment. But even with the most detailed maps, there is always a pressing need for subsequent micro-siting, analysis and due diligence on estimated energy yields. Through this process all relevant parties can learn from the wind measurements and further decrease the production risk associated with new wind developments.

Production uncertainty represents only one of the four key risk categories, each representing a potential "weak link" in the chain to sustainable wind development. In each area, the government can play a key role in levelling the playing field by making available high-quality data sets (on terrain, land use, flora and fauna, etc.) that establish a clear starting point for developers.

However, measurement campaigns have a high cost. In some cases, for example in the MENA region, competitive tendering schemes are adopted. There is a possibility that after carrying out detailed measurement campaigns, some bidders might not win the project on the site they prospected. To reduce the above risk for developers, a new concept has been introduced. A joint measurement campaign is financed by all shortlisted bidders, and carried out by international consultants. This concept was applied to the first Egyptian private wind project of 250 MW.

Download Pdf

Solar and wind resources can significantly vary spatially due to the local meteorological conditions. In order to obtain an overview of resource availability of a given region, the solar and wind energy resource is first mapped, using the available information. These maps are often incorporated into resource "atlases" that provide a detailed summary of all of the information available on the resource.; The maps are typically used to initiate policy discussions; to provide preliminary information on resource availability for pre-feasibility studies of proposed renewable energy projects. Since the amount of available energy varies significantly over time, most maps represent the information aggregated over particular stated time periods, such as long-term annual averages. For example, a wind map will typically show either the average annual wind speed or the wind power density at specified heights above the ground. The maps might also include statistical parameters (such as Weibull parameters) describing the wind speed statistics. The parameters can sometimes be displayed by direction, forming a 'wind rose'.

More information on statistical description of the wind parameters used in wind maps can be found at: www.windpower.org/en/knowledge/windpower_wiki.html

A solar resource map typically shows the total amount of solar radiation, usually in units of kWh/m2, falling on the earth's surface over a specified time frame, such as one year. Several parameters are used to quantify the solar resource: the Direct Normal Irradiation (DNI) which is applicable to concentrating collectors; the Global Horizontal Irradiation (GHI) used for flat plate collectors; and the Diffuse Horizontal Irradiation (DHI), which is the total radiation reaching a collector from the sky and clouds. The relationship linking the components is:

GHI = DNI x Cos(SZA) + DHI (1)

where SZA is the cosine of the zenith angle between the sun and a horizontal plane at the earth's surface.

More information on the parameters describing the solar irradiation can be found at: http://rredc.nrel.gov/solar/pubs/shining/chap1.html

Download Pdf

Although historically resource maps were generated from available weather data such as wind measurements or sunshine recorders, this approach resulted in large inaccuracies in the information being presented, in part due to the physical separation of individual measurement systems and non-representative data. Modern-day resource maps are the product of numerical and empirical models, which take into account a number of input parameters, including data from ground measurement networks. The underlying data are generally produced in a gridded format that can be used as a foundation for calculation of technical and economic potentials, as described in the next section, and also as a way of visualising the level and distribution of the resource in map form.

As a prerequisite, it is important to keep in mind that outputs from models used to produce resource maps are dependent on the quality of the input parameters. With modern computing techniques, it is possible to generate maps with a spatial resolution of the order of 0.1 km x 0.1 km. However in most cases, if the input parameters are incomplete or of low quality, such spatial precision is not appropriate or even possible. The final result would be misleading by its apparent level of detail, and the predictably large but unquantifiable level of uncertainty.

Wind Energy Resource Maps

Wind resource maps at a national scale or larger rely primarily on the use of outputs from numerical weather prediction models of atmospheric flow and processes run at a mesoscale level. By combining the mesoscale models with microscale models one can develop maps at extremely high spatial resolution (100 m), depending on the underlying description of the terrain (Digital Elevation Model -- DEM, and surface roughness length description), the quality of the input data sets, and the capabilities of the numerical models. The spatial resolution of the final product is greatly determined by the intended use of the model output.

Solar energy resource maps

Estimates of the solar resource over large areas can be derived from satellite images, using the visible-channel sensors in weather satellites (in particular, geostationary satellites such as the European Meteosat satellites or the U.S. Geospatial Operational Environmental Satellites (GOES), but also, to some extent, sun-synchronous polar orbiting satellites), which obtain high spatial resolution imagery of the earth-atmosphere system.

These visible channel satellite data can then be converted to high-resolution ground-level solar resource estimates using physical or empirical models, and account for such natural phenomena as cloud characteristics, atmospheric aerosols, and the influence of terrain. A detailed discussion can be found in (Renneet al, 1998)

For example, the NASA Surface Solar Energy dataset provides global coverage of satellite-derived surface solar energy at 100-km resolution using a physical-based model of radiative transfer processes through the atmosphere from the ground to the satellite. Higher-resolution data from individual geostationary satellites are being developed routinely by a number of organisations using empirical models. These empirical models are based on a "cloud index", which relates the clear-sky ground reflectance (as "seen" by the satellite) to the reflectance when clouds are present, which is almost always brighter at the satellite than the reflectance from the ground under clear skies. The figure below illustrates the computation of the cloud index used in these empirical models.

Figure 1: Solar resource assessment based on satellite images. The left image shows a typical visible satellite image over the Arabian Peninsula. Clouds are almost white, land is gray, water almost black. The middle image shows a reference image containing no clouds, just the ground information. By calculating the differences pixel by pixel between the first and second image, the cloud information can be extracted. The right picture only contains the cloud information. A white pixel is associated with a cloud index of 1 and low transparency for solar radiation, a black pixel is associated with a cloud index of 0 and full transparency for the incoming solar radiation. The images are recorded at high spatial and temporal resolution (0.10 and 15 to 30 minutes), allowing the solar radiation reaching the earth to be computed for each image at a high temporal and spatial resolution.

Satellite-derived estimates need to be benchmarked with high quality ground measurement stations in order to reduce the risk of bias or systematic error. Within the field of solar energy systematic methods for benchmarking resource data sets have been developed within the International Energy Agency (IEA) Solar Heating and Cooling Task 36 and the EU-funded project MESoR. The protocol from these benchmarking schemes should be followed to make comparable validations.

Since the underlying meteorological phenomena have different timescales (for example decadal cycles that contribute to inter annual variability and other climatic phenomena, seasonal variability, diurnal cycles and short-term fluctuations), the longer the time series considered for generating the map, the morevariability is captured at the different timescales reflecting the true average energy available.

An example of the effects of inter-annual variability on GHI and DNI is presented in Figure 2. The first graph shows a time series of annual sums of GHI in Potsdam, Germany. The blue line is the relative annual deviation compared to the long-term average from this time series, and is taken as reference 100%. The purple line is a moving 10-year average.

The second graph shows the maximum deviation of different averaging periods, it additionally contains a time series of GHI as well as DNI from Eugene (USA). The first set of bars (red, green and blue) shows the maximum positive and negative deviations of the annual values. The second set of bars shows the deviations of two year averages, the third of three year average and so on. The variability is caused by changing weather patterns from year to year; the graphs do not contain any measurement uncertainty. In the top graph for Potsdam, the individual annual GHI values show a strong deviation around the long-term mean of about 15%. The 10-year moving average is already much smoother and shows lower maximum deviations. The lower graph compares the different averaging periods to the total average of the time series. Inter-annual variability of DNI is typically higher than for GHI. The graphs show that for a given year of measurements to be within 5% of the long term GHI value, an evaluation of at least ten consecutive years of resource data is necessary. Figure 2: Example of annual variability of solar radiation.

Data quality benchmarking tool

A large amount of knowledge is available in the field of solar resource mapping. Several datasets exist, computed using various methodologies. For the end-user, then, it may be necessary to assist in identifying the appropriate existing dataset for a given application. A useful on-line tool that describes a number of solar data sets and their application toward specific uses has been developed by the United Nations Environment Programme (UNEP, 2012). The tool enables the end-user to access the description of the existing solar datasets, depending on the location or the final use for the information: www.unep.org/climatechange/mitigation/RenewableEnergy/SolarDataset/tabid/52005/Default.aspx

Download Pdf

Resource measurements, either those derived from historical sources such as national weather service observations, or obtained specifically for renewable energy programmes, are absolutely essential in the resource mapping and the renewable energy deployment process. Especially when moving into the project feasibility, due-diligence, and design and operation phases, high-quality in-situ resource measurements are essential for obtaining low-risk financing and for effective system operations. Nearby weather station data or model-derived data are no longer sufficient for these stages of development. Thus, these next two sections provide background information on how to collect high quality in-situ wind and solar measurements to supplement (and perhaps to validate) resource maps, which are primarily derived from modelling estimates.

Wind Resource Measurements

For an appropriate assessment of a wind farm, accurate on-site measurements of wind speed and direction characteristics at or close to the proposed hub height of the wind turbines are critically important. The same applies to turbine performance assessments, because of the variation in wind speed with height above the ground , and that turbine power production is proportional to the cube of the wind speed, accurate instrument specifications have been established (IEC 61400-12-1) to measure wind resources. The proper installation of instruments on a mast, and instrument calibration and recalibration are very important aspects to ensure the quality of the wind measurements. An International Network for Harmonised and Recognised Measurements in Wind Energy, or MEASNET, has been established as evidence and recognition of the high value that must be attached to exacting standards and practices in wind measurement.

MEASNET is a collaboration of companies that are engaged in the field of wind energy and want to ensure high quality measurements, uniform interpretation of standards and recommendations as well as the interchangeable value of results. The members established an organisational structure for MEASNET and perform mutual and periodic quality assessments for their harmonised measurements and evaluations. www.measnet.org

In most mapping studies, very few measurement systems are available through national weather services that meet these specifications. For example, a standard height above the ground for most weather-related wind speed measurements is 10-m, and in many cases masts may be as low as 6-m above the ground, well below the hub height of some large operational wind turbines.

Weather services often collect wind speed and direction data at airport locations and in some cases even at weather offices located in the middle of urban areas, which are typically not representative locations for high wind resource regions. Thus, for validation of wind resource maps, it is essential that ground measurement data at potentially high wind resource regimes using the IEC 61400-12-1 measurement protocols be followed to assure that wind maps derived from means such as numerical weather prediction models are properly validated.

For this reason, most national resource assessment campaigns benefit from a first analysis using a meteorological modeling approach, with validation at any existing measurement locations. The optimum location of the measurement masts depends on the characteristics of the model that will be used to generate the final map. It is therefore not always the best solution to implement masts in locations where high winds are expected. Some models require measurements in different climates or terrain conditions, to better capture the local effects.

Solar Resource Measurements

Historical measurements of the solar resource are generally available from government-supported institutions such as weather observation offices. The solar measurements are very sensitive to the frequency and the maintenancequality. The experience shows that measurements from weather services usually contain gaps and ground measurement datasets require a careful quality assessment before they can be used reliably in a solar resource assessment programme.

Historically, solar measurements were often obtained in the form of observations of the number of direct sunshine hours, using simple instruments such as Campbell-Stokes recorders, which are essentially glass spheres which burn a piece of paper above a certain level of direct solar radiation; for those times when the sun is obscured by clouds the burn marks cease to occur. At the end of each day the weather station observer can determine the number of hours of direct sunshine by measuring the total length of the burn marks.

Many weather services have also undertaken actual solar radiation measurements, using instruments such as thermopile-based or silicon-diode based pyranometers. Thermopile based measurement have the advantage to be spectrally neutral, while measurements based on silicon sensor can vary with the cloud situations and atmospheric conditions, or the sensor temperature. In some cases weather stations might also operate instruments that obtain the direct normal component of the solar resource, using either a tracking pyrheliometer that follows the solar disk across the sky throughout the day, or a shading-type system that allows for the determination of the diffuse sky radiation. These instruments are very sensible to a misalignment of the tracking system and need careful maintenance during their operation, to ensure data quality.

Solar measurement databases

Among the best sources of information for obtaining quality solar measurements are publications developed under the Baseline Surface Radiation Network (BSRN) programme² and a recent publication by National Renewable Energy Laboratory (NREL) (Stoffel, et al., 2009). The quality and completeness of measured data from networks is an important element to take into consideration before using the information to develop resource maps and for project planning.

• In the U.S. National Solar Radiation Data Base (NSRDB) and the Typical Meteorological Year data derived from this database, three classes of data quality from networks are defined: Class I, which have the lowest uncertainty data; Class II have higher uncertainty; and Class III which have incomplete data records. The type of instrument used, and the calibration and maintenance procedures followed in the measurement programme typically define the data uncertainty. Historically, well-calibrated and well-maintained thermopile-type solar measurements are considered to provide the lowest uncertainty, but are also the most expensive to operate. Higher uncertainty comes in the use of more poorly calibrated and maintained instruments. Incomplete data records are the result of instrument failures during a portion of the measurement period. More details can be found in Wilcox and Marion (2008).
• Ground measured data are available at different temporal resolutions. The high quality Baseline Surface Radiation Network (BSRN ) provides 1 minute values, but only for a very few location around the world.
• Many weather services record hourly values, but they are often not publicly available. Many long-term archives are available through the World Radiation Data Center (WRDC), but only store monthly values.

Despite the availability of national network data from around the world, many governments and private companies undertake additional solar measurement programmes. Although these programmes are undertaken for research purposes or other purposes outside of the energy field, more and more measurements are being undertaken specifically to support the development of solar energy projects and to provide further validation of national solar resource maps. A definitive report on best practices for conducting such measurement programmes and the accuracy that one might expect from measurement campaigns using various instrument types of field practices was recently published by NREL (Stoffel et al., 2009).

Figure 3: A high quality solar radiation measurement station with a sun tracker. The left unshadedpyranometer records the global horizontal radiation, the pyrheliometer (the small tube) the direct normal radiation. The pyrheliometer is always pointing towards the sun. The right pyranometer is shaded from direct sunlight by a small disk. This pyranometer therefore only records the diffuse radiation. A small misalignment of the tracking device will make the direct and diffuse measurement fail as they do not point to the sun anymore or if the pyranometer becomes unshaded.

Download Pdf

Quality resource maps provide a critical overview of resource availability and distribution over a region, and the foundation for preliminary project planning and pre-feasibility assessment. Resource maps, or, more specifically, the underlying gridded data that make up the maps, also provide the basis for calculating the technical and economic renewable energy potential for a given region. Calculating these potentials is a key next step to take once a resource map is constructed. Knowledge of their renewable energy potential informs national as well as global policy discussions, since they provide a quantitative basis for renewable energy deployment opportunities and development scenarios.

It is important here to note the difference in definition of potentials for renewable energy, compared to fixed stock resources. The total resource base of fixed stock resources (including fossil fuels such as coal and oil) can be subdivided according to the McElvey Diagram (USGS Bulletin 1450-A), shown in 4. In this figure, fixed stock reserves are subdivided into "proven" and "estimated" categories, and further subdivisions are performed based on the accessibility of the resource and the economic certainty in the ability to extract the resource. As such, "proven" reserves are generally much smaller than the total resource base, and the quantification of this subdivision has important implications on decisions for extracting and pricing these fixed stock resources, since the market price depends on the ability to supply the demand on the long term.

Figure 4: A simplified example of the McKelvey Diagram (from Maxwell and Renne, 2004, adapted from USGS Circular 1450)

Wind and solar resources are "flow" resources and exist as the propagation of energy (Maxwell and Renne, 1994). Those resources have three main characteristics that distinguish them significantly from fixed stock resources: 1) they are temporally variable, 2) their consumption does not alter the flow or deplete them at the location they are being used, and 3) the resource varies over time, and does not depend on the demand.

For these reasons specific definitions were adopted for renewable energy 'potentials'. While the resource represents the flow, or the 'amount of' sun or wind at a given place, described by an energy flux, the 'potential' provides a realistic estimate of the portion of this flow that can be used for energy applications; theoretically, technically, or economically. "Potential" is formulated in energy terms - electricity or heat.

An attempt has been made by NREL to illustrate the various aspects of renewable energy potentials. One can discriminate the theoretical potential from the technical potential and the economic potential (5), explained in the following sections.

Figure 5: A conceptual diagram of Renewable Energy Potentials (from NREL, 2012)

Download Pdf

The Intergovernmental Panel on Climate Change IPCC (2012) defines the theoretical potential as the amount of the physical energy flow that could potentially be used at a specific site and during a specific period (e.g. over the course of one entire year).

Examples are the total amount of solar irradiation, the kinetic energy of wind or the gravitational potential energy of water. Due to the fluctuating nature of many renewable resources, long-term averages are most suitable to derive the theoretical potentials (Stetter et al., 2011). Thus, the theoretical potential represents the upper limit of energy usable for human applications that can be produced from an energy resource based on physical principles and current scientific knowledge. The theoretical potential does not take into account energy losses during the conversion process necessary to make use of the resource, nor any kind of technological or geographical restrictions.

On a global scale, the estimates of theoretical potential for wind and solar technologies vary widely but in all studies the theoretical potential far exceeds current global energy demands. For example, the IPCC (2012) report indicates a global theoretical solar potential of 3.9 x 10⁶ EJ/y (1.08 x 10⁹ TWh/y). For wind, the global annual flux is about 6 x 10³ EJ/y (1.67 x 10⁶ TWh/y). By contrast, the annual global energy consumption in 2010 is estimated in BP Statistical Review of World Energy (2011) to be 5.28 x 10⁴ TWh/y.

Although the IPCC (2012) states that the information on theoretical potentials has limited practical relevance, theoretical potentials do provide an important policy-relevant perspective, as it is clear that renewable energy potential is not at all a limiting factor on a global scale for achieving significant carbon emission reductions using renewable energy technologies.

Portraying the theoretical potential in map form provides critical background information on where the most favourable resources might be located. At regional and country levels, maps of theoretical potential can also be used to stimulate policy discussions, particularly if they present evidence that renewable energy resources might be sufficient in their country to warrant these discussions. The resource maps described in Step 1 generally depict the theoretical potential. As a matter of caution, however, one might keep in mind that the locations with the most favourable resources are not always technically feasible, or the most economically feasible for project development. For this reason the concept and definition of technical and economic potentials have been established, and are discussed in the next sections.

Download Pdf

The technical potential is defined by IPCC (2012) as the amount of renewable energy obtainable by full implementation of demonstrated technologies or practices. This definition does not detail the notion of 'full implementation' and 'demonstrated technologies', nor 'practices'. From the literature, the basic principle to assess technical potentials is to restrict the available land area to take into account constraints created by competing uses, excluded areas (such as national parks, water bodies, and other protected areas), inadequate locations due to technical constraints, environmental constraints such as bird migration paths, and other restrictions.

Once the suitable land areas are identified, the resource estimate on those locations is used to compute energy estimates, using technical assumptions. The final result can be presented as a single value - the sum of the expected production, or in the form of a map. In the following sections, the literature is studied in detail in order to provide a list of the most commonly identified constraints and technical assumptions.

There is no agreed definition of the parameters needed to evaluate technical potentials, leading to significant differences in approaches, and final results. Two major approaches emerge from the literature analysis. For the most common approach, a Geographic Information System (GIS) is used to superpose information layers over a Digital Elevation Model (DEM) and a resource map, creating exclusion zones. The other provides an estimate of the suitability of each location, by ranking each criterion based on a point system.

Download Pdf

A major lesson learnt from the literature assessment is the need for cautionin evaluating technical potentials. Increasingly accurate resource maps are being provided both by public and private sector entities. The use of Geographic Information Systems (GIS) enables the use of rapidly superimposed layers of information to develop apparently definitive numbers on the technical potential.

The more precise the resource map, the greater might be the temptation to superimpose all possible restriction layers to identify 'suitable areas'. Each parameter added to the analysis, however represents an additional assumption, whose impact needs to be carefully considered. For example, by including the distance to the grid, the underlying assumption is that the project economics would not enable projects to be built beyond a certain distance. However, the economics of a project depend on several factors, besides the resource and distance from the grid, such as project size, intensity of the tariff or support scheme and the costs of extending the grid to the proposed development site.

Fortunately, the ability to generate detailed resource maps comes with the computational ability to handle large amounts of related information, and to develop flexible analysis tools.

As each location has its unique characteristics, there is no single method that can be adapted explicitly to the whole world without introducing uncertainties.

Important recommendations for determining technical potentials:

There is no single, definite value of technical potentials. Technical potentials fall within a range of values, depending on the parameters considered, and the uncertainty on the input parameters and models.

• As a consequence, a map of technical potentials should detail the assumptions used to build it, and perform a sensitivity analysis to the input values.

Approaches based on exclusion zones should be used carefully. One should discriminate locations where developments are legally prohibited, or technically impossible, from locations where developments are unlikely to happen. With modern GIS systems, the quality of a location can be flagged by putting an indicative weight to the constraints.

Given the resource-sensitive nature of wind technologies, coarse-resolution (50-100 km) maps are likely to introduce large errors into the calculation of theoretical and technical potentials, possibly significantly underestimating the resource. The same is true for resource-sensitive solar technologies, such as CSP. Adding information layers to limit the available land area does not compensate for the high uncertainty on the resource map and does not lead to more accurate values. Therefore, studies using coarse-resolution resource data are likely to have less value for supporting policy and investment decisions than studies based on higher-resolution data.

Authorities in charge of planning developments and project developers are more likely to use more detailed maps when evaluating project feasibility. However, there are uncertainties associated with the input data sets, and these uncertainties may override the accuracy and precision of the resource data. Although maps of technical resources might highlight opportunities, ground-truth assessments will help validate the analysis and lead to more "bankable" data sets for project investments.

Areas identified as unlikely for future developments might become suitable in case of technology upgrades, or changes in policies or economics. An assessment of the technical potentials is a living project, which needs to be revised regularly.

Therefore, an analysis of technical potentials can be used as guidance, but not as a prescription for where development can occur. Technical potentials provide an overview of the possible location of areas of interest. This information is useful at a high level to estimate the possible contribution of the technology to the energy mix of the country, pre-identify areas for further exploration through measurement campaigns, initiate the involvement with local communities, and to run simulation models to investigate the behaviour of the electricity system when integrating different shares of renewable energy.

Download Pdf

Exclusion-based approaches

There are significant differences found in the literature in approaches between analyses at the global level, using coarse resolution data and analyses carried at the regional or country level.

Many of the global assessments (as in Hoogwijk, 2004, and later revisions) were based on wind data at low spatial resolution, from the Climatic Research Unit (CRU) of the University of East Anglia. The CRU data provide worldwide climatic average (1961-1990) measured values from measurement stations interpolated on a regular grid of 0.5 x 0.5 degrees (approx. 50 x 50 km). Other approaches for estimating global wind resource have been made using reanalysis data such as GEOS-5 DAS (Lu et al., 2008), with a resolution of 0.67 x 0.5 degrees, CRU and reanalysis datasets are characterised by a relative coarse resolution.

However, the latest research performed by Risoe/DTU (Badger and Jorgensen, 2011) demonstrates that using coarse resolution datasets leads to a significant underestimate of the resource. The reason for this is that a significant amount of energy from the wind is present at smaller scales, which is not captured by these coarse datasets. This latest research might lead to the conclusion that up to now the global estimates of wind resource significantly underestimate the total wind resource.

At the national or regional level, wind resource estimates are based on resolutions ranging from less than 5 x 5 km down to 0.2 x 0.2 km. To obtain such detailed results, downscaling methods are used. These methods generally require additional models, which use as input the output of the publicly available reanalysis datasets. These additional models are generally referred to as mesoscale models, and are used to downscale the reanalysis data to 50 km resolution or less. In many cases, "nested" mesoscale models are used to further downscale to 5 km resolution or less.

Models used in downscaling methods include the microscale models within WAsP (DTU Wind Energy's Wind Atlas Analysis and Application Program), the mesoscale models MC2 (Mesoscale Compressible Community model), and WRF (Weather Research and Forecasting (WRF). The output requires validation against ground measurement campaigns, to assess the uncertainties in the model outputs.

Due to the inter-annual variability of the resource, the temporal coverage must be sufficient so the results are not biased by using data collected only during high- or low-resource periods. For this reason, a multiple number of complete years of data must be used, so that all seasons are equally represented. Today, some reanalysis data sets are available for more than a 30-year period, a period that is deemed to be climatologically significant.

These wind resource data sets form the basis for the calculation of technical potential. Technical potential analyses carried at 0.5⁰ spatial resolution are based on a 'suitability factor', applied to the different categories of land use and land cover. These suitability factors are defined as percentages of land area of a given wind speed class that can be used for generating power. Elevations above 2000m to 2500m are excluded in most cases . The suitable areas are limited to average wind speeds >4 m/s at 10 m height. The remaining land surface is multiplied by a maximum capacity per land unit surface (>4 MW/km2), in order to estimate the maximum installable capacity. The core data layers for such analysis include the land use, land cover, topography, and the resource map.

For analyses carried at spatial resolutions of 5 km and below, additional data layers are superposed over the resource map, using Geographic Information System (GIS). These datasets are used to refine the 'suitability factor' based on a more realistic estimate of the available land.

• In most studies, the distance to the grid is a key parameter, and a maximum allowable distance is set. Depending on the study, the distance varies from 15 km to 75 km, either as a geometric distance to the closest transmission line, or to the nearest substation.
• The allowable terrain slope is set to a maximum of 15 degrees to 20 degrees,
• Elevations above 2500 m are excluded,
• Lakes and rivers are excluded.

Various limitations are applied on a case-by-case basis:

• The most common is the distance to settlements, with minimum distance varying from 400 m to 1 km, and a maximum distance from 8 km to 50 km.
• In two cases, the distance to roads is considered, with a minimum distance of 0.1 km, and a maximum distance of 80 km.
• Stetter et al., (2011) and Schillings et al., (2009) include the distance, not only to the grid but also between the power plant and load centres.
• Other limitations consist of: protected areas, specific geological conditions, urban areas or military areas.

The suitable areas are further limited to average wind speeds >6-7 m/s at 50 m height. The remaining land surface is multiplied by a maximum capacity per land unit surface (>5 MW/km2), in order to estimate the maximum installable capacity.

Approaches by exclusion parameters have strong limitations. In particular, some of the parameters taken into account are not proper exclusion parameters, but an attempt from the authors to set reasonable limits, or to exclude unlikely developments. On the other hand, some suitable areas may actually be unsuitable for project development if some other factors are overlooked.

Opportunity-based approaches

A different approach, adopted by Schilling et al., (2009), ranks the quality of sites by assigning a score to each major dimensioning parameter. The combination of the scores provides the final ranking of the site.

In Schilling et al., (2009), the criteria taken into account are: the resource, distance to transmission lines, distance to network nodes, roads, and load demand. The result is displayed as a map presenting a qualitative ranking for all potential sites.

This approach is of great interest because it does not intend to superimpose exclusion layers but combines the different technical parameters to assess the likelihood or 'probability' of projects that can occur in the considered area.

Example: With a rigid approach by exclusion zones, a project like the Turkana wind project would have been excluded from consideration, representing a missed opportunity. This project of 300 MW is situated in northwest Kenya, an area of very high wind speeds, which makes the project technically feasible, although far removed from transmission or transportation access. A 428 km transmission will be built for the purpose of connecting the wind farm to the electricity system, and a large section of road will be reinforced for the purpose of the project.

Estimating the extractible wind energy

Once the land availability has been determined, it is possible to evaluate the energy content - the technical potential. For wind, the calculation is complex since wind turbines cannot produce any energy at all below certain winds speeds, turbines cannot be spaced in such a way as to capture all of the horizontal wind flux, and the conversion efficiency is limited by the Betz limit (0.59 of the incoming flow).

Several methods exist, based on classes of wind speed, or using virtual or real power curves of wind turbines.

One method used by NREL, is based on the concept of wind power classes to specify the wind energy resource at a given location. Each location falls in a specific class, numbered from 1 to 7, which corresponds to a level of power density. The power density is itself based on the wind distribution statistics (often assumed to follow a Weibull distribution), built from hourly data over a long time series. Table 2 provides the results of a hypothetical example of the methodology described below.

The approach is implemented as follows: 1) assume an installed capacity of 5 MW/km2 of "windy" land (land assumed of sufficiently high wind resource to support current wind turbine technologies); 2) Assume that the wind power class must be 3 or higher to qualify as "windy" lands, and 3) assume a wind power conversion to energy production for each wind power class (higher values for higher power classes). The technical installed capacity for wind is defined as:

(Technical Installed Capacity)wind = 5 MW x ∑ A_c

Where Ac represents the total area of land falling within a wind power class, c, where c > 2. Then:

(Annual Technical Potential)wind = 5 MW x 8766h ∑ (C_(p,c) x A_c)?

Where Cp,c = the capacity factor for a wind turbine for each power class. For this example, the following table shows how the technical potential for a 100 km2 land mass with spatial variability in the resource would be made.

This method is sensitive to the facts that 1) even low wind speeds offer some theoretical wind power potential (even though the technical potential is zero); 2) it is based on wind classes and does not look at individual wind speeds; and 3) the installed capacity assumptions are somewhat arbitrary and not necessarily reflective of the possible maximum installed capacity.

Table 2. Hypothetical wind power class and related power coefficients for a 100 km2 land mass (assume installed capacity = 5 MW/km²).

Download Pdf

Most assessments performed by the literature at a global level were mostly carried with datasets with a spatial resolution of 0.5 x 0.5 degree (approximately 50 x 50 km), using NASA and CRU data . At the country level, estimates were performed down to 5 x 5 km or lower. The analyses differ significantly between solar photovoltaic (PV) and concentrated solar power (CSP).

For solar PV, urban areas are not excluded when considering rooftop applications. The land availability is constrained mostly by considering protected areas, and land occupation such as agricultural and forestry areas, grasslands, or any area assumed unsuitable due to socio-geographical reasons. Other parameters include the spacing between rows of PV systems to avoid shadowing of adjacent systems. In two studies no land area is excluded and no specific spacing factor between modules is considered.

For solar CSP, the analyses are similar to wind energy, and mostly based on exclusion criteria, except with the study bySchilling et al., (2009) where ranking criteria has been also applied for CSP evaluation. Those similarities with wind energy exist because both types of installations have a large footprint. Land slope is however an important constraint within CSP studies. There are strong similarities of analyses based on low (50 km) and relatively high-resolution resource datasets (1 km -10 km).

Common dimensioning parameters for CSP are:
• A minimum level of Direct Normal Irradiation (DNI) between 1445 kWh/m²/y and 2400 kWh/m²/y,
• The distance to lake or rivers ,
• A maximum slope fixed between 2.1 degrees and 7 degrees.
• In most studies, the distance to the grid or nearest substation is taken into account,
• Urban areas are systematically excluded, with some consideration of distance to settlements (minimum 1 km, maximum 50 km).

Additional exclusion parameters for CSP are considered:
• Similar to PV, protected areas and land use constraints are excluded. In one study, a buffer of 10 km is considered from sand dunes or glaciers.
• Some studies take into account the competition for land area or rooftop locations among the different solar technologies (PV, solar water heating, CSP) and with other renewable technologies, such as wind.

Estimating the extractible solar PV energy

Once the land availability has been established, it is possible to evaluate the energy content - the technical potential.

The rating for PV systems, for example, applies to system output rating at a resource level of 1000 W/m2. Thus, for a 100 km2 area, the technical installed capacity for PV systems that operate at 10% conversion efficiency is 10 GW (10,000 MW), as per the formula:

(Technical installed capacity)solar = Area x system rating condition x conversion efficiency [3]

The Technical Energy Potential for this condition is then:

(Technical Potential)solar = (Technical Installed Capacity/system rating condition) x resource = Area x conversion efficiency x resource

Thus, the technical energy potential for the installed capacity described above and for a 100 km2 land mass with an average resource of 5 kWh/m²/d is 50 GWh/d or 18,250 GWh/y.

Download Pdf

GIS tools have been used by the National Renewable Energy Laboratory (NREL) to develop estimates of the CSP technical potential in the south-western U.S., and highlighting the most favourable regions for CSP development (Mehos and Perez, 2006). The study was done in part to support a renewable energy feasibility study for the U.S. Bureau of Land Management (BLM), a major federal government land manager with significant land holdings under its mandate in the western U.S. (Heimiller, et al., 2003).

The study begins with a high-resolution, high quality satellite-derived DNI resource map at 10 km spatial resolution for the region (6a). From this map (which represents the total DNI and therefore the theoretical resource for the region), a variety of GIS "layers" were applied to the data to remove regions where CSP development could not occur. These layers are based on the following exclusion criteria:

• Eliminate all regions where the DNI solar resource >2190 kWh/m²/y
• Remove all lands with environmental and land use exclusions (such as national parks, urban areas, sacred grounds)
• Remove all land area where the terrain slope is >1% (CSP technologies require extremely flat land).

The result of this simple screening process is shown in 6b. Even with these restrictions, the technical CSP potential for the region is extremely high (>26 million GWh/y).

For the BLM study additional restrictions were applied:

• Potential project sites had to be within 80 km of 115 kV-345 kV transmission lines;
• Potential project sites are within 80 km of a road or railroad;
• A minimum parcel size of 40 continuous acres must be available;
• Land use is BLM-compatible

The results of the BLM analysis provided the agency with sufficient information to develop rulings for renewable energy development on their lands (using a leasing scheme). Figure 6: (left) Theoretical DNI resource potential in the western United States, as derived from geostationary satellite data. (right) Areas suitable for CSP development, based on DNI resources >6 kWh/m2/d, (2190 Kwh/m2/y) after exclusion factors 1-3 (see text) have been applied.

Download Pdf

A study has been conducted by the CENER to analyse the CSP potential using GIS and solar data validated against DNI measurements available in Africa. For this analysis different GIS layers have been considered, these layers are based on the following exclusion criteria:

Eliminate all regions where the DNI solar resource < 1800kWh/m2/y
2. Remove all lands with environmental and land use exclusions (such as orography, water areas, environmental zones, transmission lines and feasible areas)

Figure 7: (left) Theoretical DNI resource potential in the ECOWAS, as derived from NWP modeling. (right) Areas suitable for CSP development, based on DNI resources >1800 kWh/m2/y, after exclusion factors 2 (see text) have been applied.

Download Pdf

The Economic Potentials take into account all social costs, assume perfect information and the market potential of renewable energy sources, that depends upon existing and expected real-world market conditions shaped by policies, availability of capital and other factors(IPCC, 2012).

The full definition of Economic Potential (from IPCC, 2012 Annex I) is the amount of renewable energy output projected when all social costs and benefits related to that output are included, there is full transparency of information, and assuming exchanges in the economy install a general equilibrium characterised by spatial and temporal efficiency. Negative externalities and co-benefits of all energy uses and of other economic activities are priced. Social discount rates balance the interests of consecutive human generations.

Based on a map of technical potentials, it is nowadays technically possible to run a model on each identified location to assess different economic or financial parameters.

However, considering the source of uncertainties both on the resource and the technical potentials, generating maps of economic potentials over a large area could lead to extremely misleading results. Such analyses should, in general, be conducted over limited areas, based on datasets validated against ground measurements, and provide a detailed justification for each entry parameter, as well as a sensitivity analysis. For example, the discount rate used to calculate the levelized cost of electricity is often a subject for debate in the modelling community, due to its large influence on the final results.

Intense research is on-going in this field, since the calculation of economic potentials is a powerful tool for the promotion of renewable energies. The estimate of the economic potentials requires transparent and documented methodologies and depending on the selected input parameters or model. The estimate can be used to:

• Compare the levelized cost of electricity with current generation solutions. In some circumstances, renewable energy technologies are already cost-competitive with traditional power generation options, in particular when externalities are included, as described by the IPCC definition previously mentioned.
• Estimate the impact of a support scheme on the deployment of a technology, and map where the development might happen, given a minimum profit margin. On this basis, and selecting the economically feasible areas, it is possible to run pre-feasibility analysis on the ability of integrating renewable generation into the grid, or developing scenarios of the energy mix evolution, as described in the next section.
• Disclose information about the economic feasibility and its uncertainty - for investors, which increase the attractiveness of the market to the financing community and the industry for further market evaluation.

There is not yet a consensus for assigning a value to the 'negative externalities and co-benefits'. A range of models and methods is available, which fulfils partly the IPCC definition. Each model uses its own entry parameters and internal calculations, which make the outcomes difficult to compare. A summary of the dimensioning parameters considered by the literature to estimate the economic potentials is provided by the below tables.

When calculating the economic potential of renewable plants several parameters must be considered as input such as investment cost, interest rate, power plant lifetime and operation and maintenance costs. Those inputs are used in widely spread financial models such as Retscreen , SAM, ReEDS or WINDS that calculate the economic viability of technology deployments.

The estimation of the economic potential often varies from one approach to another; this is mainly due to different assumptions. Major dissimilarities have been observed in the interest ratevalues, electricity cost cut-off prices or inflation rates. Other assumptions that complicate the comparison of economic results are time value of money and currency, growth rates or even technology development assumptions.

The present and future economic viability of projects is tightly related to stable renewable tariffs and power targets along with public promotion and incentives.

The importance of correct policies and incentives ensure the economic viability of renewable energy projects.

Solar - literature analysis of the dimensioning parameters for assessing the economic potentials.

Wind - literature analysis of the dimensioning parameters for assessing the economic potentials

Download Pdf

The Spanish government together with the IDEA (Institutoparala DiversificacionAhorro de la Energia) made a financial-economic analysis of the national renewable investment plan determining the technical economical parameters for each type of technology considered. The parameters considered are the following:

• Time period evaluated, yearly based
• Time period of execution, estimated time to undertake and execute each type project
• Installations lifetime, based on average values
• Equivalent operation hours, based on each technology known experience
• Expected production, evaluated with operational equivalent hours and installed power of each project

Energy selling price, for projects with selling energy option is estimated with the feed-in tariff or the market price with bonus weighted with an IPC parameter (Price Consumption Index). For projects for replace and/or saving energy, the substitute energy cost per energy unit equivalent

Material investment cost, corresponding to fixed assets during the whole execution period

Operational and maintenance (O&M) costs, based on the necessary expenses to ensure the correct functioning of the power plant

Price Consumption Index (IPC), based on the updating factor which has been considered as steady (2%) during the evaluation period

Mean Electrical Energy Price Index, actualisation factor considered constant (1.4%) during the evaluation period

Companies' tax, fixed to 35% and constant along the time period evaluated

Project profitability, estimated on the Investment Rate of Return (IRR)

- Income statement, with the previous assumptions a predicted income statement is calculated to provide the project's profitability through the cash flow calculation

- For calculating the costs the main parameters considered were the investment cost, interest rate, power plant lifetime, and operation and maintenance costs.

Download Pdf

The analyses on the theoretical, technical or economic potentials are performed for a specific time (present), and provide an estimate of the maximum expected deployment in given conditions. However, technology deployment does not happen overnight and a timeline has to be developed to schedule this deployment, expand the infrastructures, and mobilise investments and civil society.

Scenarios are used to assess the consequences of an energy policy, and perform retro-active planning exercises to initiate the concrete steps for the development. As shown in figure 9, those concrete steps include policy creation, market development strategies, and mobilising investments. Scenarios are therefore central elements of the policy debate, be it at global level (IPCC, IEA, Greenpeace, and many others), or even a national or regional level.

Building a scenario implies taking into account realistic estimates of the technological development over time, as well as other major elements like evolutions in society, market dynamics, or even energy dependence issues. Taken on this basis, a scenario is an expected installed capacity for a given timeline, and the associated energy production.

Example: direct use of scenarios in the policy-making process. A large scenario exercise was supported by the European Union for the implementation of the 2009/28/EC directive on the promotion of renewable energy use from renewable sources. The jointly agreed binding target of a 20% share of renewable sources in the overall energy consumption was divided into individual targets for each of the 27 countries, with a trajectory for deployment for each technology.

Bottoms up energy-economic models, integrated assessment models (IAMs) and computable general equilibrium models (CGEs) are used to help address many pressing questions of our time, including climate change, use of natural resources, energy and economics, international trade for a single country, and regional and global policy and technology analysis. Most of the global models are based on regional aggregation of energy economies (perhaps down to country level) and have evolved over the past decades to have relatively robust capabilities for addressing multiple technical options and feedback with environmental systems. These models are the benchmark in energy economics for climate mitigation analysis. They form the core analysis for IPCC stabilisation scenario assessments and continue to be instrumental in informing national and international dialogue and policy regarding the options and costs for mitigating climate change.

Many renewable mitigation options, including wind, solar, geothermal, and biomass, are currently poorly represented in IAM and CGE modelling. These technologies are site specific - accurately depicting the economic and technological characterisation of renewables that requires location specific knowledge of potential, transmission and possible constraints. Wind, biomass and geothermal resources are all site dependent, strongly impacting the economics across states, regions, countries and the world. Consequently, IAMs and CGEs can be improved by developing complementary, fine scale, analysis of these site dependent characteristics.

The first step in assessing the mitigation potential and economic costs of the different mitigation options is to assess the quantity and cost of supply in as much detail as possible. For the renewable energy technologies, which are location dependent, global data sets only have been used which represent features at a coarse scale of e.g. 100km - 200km. This is in contrast to the 20 m-200 m resolution required for accurate on-shore wind forecasting, to account for terrain features, as an example, and other renewable energy technologies.

Download Pdf

The process of moving from technology scenarios to investments through the creation of market strategies, government policies and financial instruments, does not follow a linear process. The process is continuous and evolves with the level of deployment of the technology. Detailing the process of policy-making and the creation of a market environment, which would highlight the main principles for using resource-based information in this process, is beyond the scope of this report.

The figure 9, is an illustration of a positive feedback cycle for PV. Policies create an enabling market environment, in return leveraging investments as part of a dynamic process. Innovation and large scale deployment leads to cost reduction of the technology, feeding into the policy development.

Figure 8: Positive feedback supporting expanded renewable energy growth: expanding markets encourage private investment, which in turn leads to manufacturing scale-up and further R&D to improve product efficiency and lower manufacturing costs. This in turn leads to cost reductions in the products. This creates an environment of further business innovation, encouraging governments to establish favourable policies that further expand the markets.

The dynamics of policy development and market creation is detailed by IEA 2008, which presents a methodology aimed at assessing the effectiveness of policies for deploying renewables, based on data about renewable energy markets and policies collected from 2000 - 2005.

The starting point of the analysis is the "realisable potential" for a given technology in the year 2020, based on renewable energy resources and technology potential for a given country. The "Realisable Potential", is thus the "Technical Potential" adjusted to take into account unavoidable medium-term constraints on the rate of deployments. In virtually all cases and in all countries the realisable potential at a given timescale far outstrips the current installed capacity.

The definition of 'effective policies' differs for each phase of the technology cycle. Technology deployment scenarios help highlight the different phases of technology maturity, and adapt policy and market instruments to stimulate their growth.

Figure 9: Framework of policy incentives as a function of technology maturity leading to the "realisable potential" (IEA, 2008)

The aim of policies and market instruments favouring the growth of renewable energy is to learn and compensate for the risks of investing in a new technology. The investment in clean energy technology will benefit the whole of the society in the long run. Financing the early stages of the market introduction from public funding is a way of sharing the development risk at the beginning among all those who will profit from it in the long run. The introduction of a new technology to the market therefore starts with a strong research, development and demonstration component, progressively evolving to controlled market conditions, until the new technology becomes competitive with those already in the market.

At an early stage, the technology is not mature, the costs are unknown, and risks on the technology and investments are too high to enable private sector investment. Government R&D programmes along with prototype testing and demonstration projects are essential to mature the technology.

As the market volumes increase, the technologies mature, costs are reduced and the risks are better evaluated. Strong market incentives enable the technology to deploy in niche markets. Effective policies therefore shift from government R&D and demonstration programmes to price-based and quantity-based incentives such as feed-in tariffs, certificates, or government-supported tenders. With increasing market volumes, and provided that continuous innovation is performed, the costs continue to decrease, resulting in "low cost-gap" technologies, such as onshore wind.

Fully mature technologies enter into mass markets. Where market deployments continue to increase toward the realisable potential for these mature technologies, policy incentives can become more technology neutral, creating a market environment that enables competition among technologies on a comparable basis.

Download Pdf

Badger, J. and H. Jorgensen (2011), "Energy Systems and Technologies for the Coming Century", Article in proceeding presented at Risx International Energy Conference 2011, Danmarks Tekniske Universitet, Risx (DK),pp215-225.

Blackwell and Veatech Coorporation (2007),"Arizona Renewable Energy Assessment", www.aps.com/_files/solarRenewable/AZRenewables.pdf.

BNEF(Bloomberg New Energy Finance) and UNEP (United Nations Environment Program) (2011). "Global Trends in Renewable Energy Financing: Analysis of Trends and Issues in the Financingof Renewable Energy", www.unep.org/pdf/BNEF_global_trends_in_renewable_energy_investment_2011_report.pdf.

BP (British Petroleum) (2011), "BP Statistical Review of World Energy",www.bp.com/liveassets/bp_internet/globalbp/globalbp_uk_english/reports_and_publications/statistical_energy_review_2011/STAGING/local_assets/pdf/statistical_review_of_world_energy_full_report_2011.pdf.

De Vries, B., D. Van Vuuren and M. Hoogwijk (2007), "Renewable energy sources: Their global potential for the first-half of the 21st century at a global level: An integrated approach", Energy Policy, 35, Elsevier, pp. 2590-2610.

Dommnguez, J., X. Garcia and I. Pinedo (2007), "GIS approach to the definition of capacity and generation ceilings of renewable energy technologies", Energy Policy, 35, 10, Elsevier, pp. 4879-4892.

Ethio Resource Group with Partners(2007),"Utilization and Project Development Scenarios" http://swera.unep.net/typo3conf/ext/metadata_tool/archive/download/EthiopiaSWERAreport_258.pdf.

Hoogwijk, M., et al.(2004),"On the global and regional potential of renewable energy sources", http://igitur-archive.library.uu.nl/12A92441-FB68-40EB-B952-241723F88C3A/FinalDownload/DownloadId-09DE36D712F401272BA636162C9719D4/12A92441-FB68-40EB-B952-241723F88C3A/dissertations/2004-0309-123617/full.pdf

Hoogwijk, M., et al.(2008), "Global Potential of Renewable Energy Sources: a literature assessment", www.ren21.net/Portals/97/documents/Publications/REN21_RE_Potentials_and_Cost_Background_document.pdf.

Hoyer-Klick, C., et al (2009), "Getting Solar Energy to Work: Resource Assessment by Remote Sensing as a Base for Investment Decisions" www.earthzine.org/2009/05/05/getting-solar-energy-work-resource-assessment-remote-sensing-base-investment-decisions/.

International Energy Agency (2008), "Deploying Renewables: Principles for Effective Policies" www.iea.org/textbase/nppdf/free/2008/deployingrenewables2008.pdf.

IPCC (Intergovernmental Panel on Climate Change)(2012), "Renewable Energy Sources and Climate Change Mitigation, Special report of the Intergovernmental panel on Climate Change", http://srren.ipcc-wg3.de/report/IPCC_SRREN_Full_Report.pdf.

Krewitt, W., Nienhaus, K., Kle_mann, C., Capone, C., Stricker, E., Graus, W., et al. (2009), "Role and Potential of Renewable Energy and Energy Efficiency for Global Energy Supply", www.ecofys.com/files/files/summary_report_role_potential_renewable_energy_efficiency_global_energy_supply_02.pdf

Lu, X., M. McElroy and J. Kiviluomac (2009), "Global potential for wind-generated electricity", www.ncbi.nlm.nih.gov/pmc/articles/PMC2700152/.

Maxwell, E. L. and D. S. Renne (1994),Measures of renewable energy resources. NREL/MP-463-6254, National Renewable Energy Laboratory, Golden, Colorado

Mehos, M. and R. Perez (2005), "Mining for Solar Resources: U.S. Southwest Provides Vast Potential",www.imagingnotes.com/go/article_free.php?mp_id=13.

New, M., M. Hulme and P. Jones (1997), "Representing Twentieth-Century SpaceTime Climate Variability. Part I: Development of a 196190 Mean Monthly Terrestrial Climatology", Journal of Climate, Vol. 12,No.3, pp 829-856

New, M., M. Hulme and P. Jones (1999)"Representing Twentieth-Century SpaceTime Climate Variability. Part II: Development of 190196 Monthly Grids of Terrestrial Surface Climate", NREL

Pletka, R and J. Finn (2009), "Western Renewable Energy Zones, Phase 1: QRA Identification Technical Repor",www.nrel.gov/docs/fy10osti/46877.pdf.

REN21 (2011)."Renewables 2001", Global Status Report.

Schillings, C .,et al. (2009), "Potential demonstration sites for wind and concentrating solar power projects and Ranked list of potential demonstration sites for wind and concentrating solar power",www.remap-ec.eu/dossier_public/REMAP-D_2_2_D2_3.pdf.

Stetter, D., and W. Krewitt (2011), " Global GIS-based inventory of renewable energy resources in high spatial and temporal resolution",www.dlr.de/tt/en/desktopdefault.aspx/tabid-2885/4422_read-16419/.

Stoffel, T.,, et al. (2010), "Concentrating Solar Power: Best Practices Handbook for the Collection and Use of Solar Resource Data",www.nrel.gov/docs/fy10osti/47465.pdf.

U.S. Dept. of the Interior BLM and U.S. Dept. of Energy EERE(2003), "Assessing the potential for Renewable Energy on Public Lands", www.nrel.gov/docs/fy03osti/33530.pdf.

Wilcox, S. andW. Marion. (2008), "Users Manual for TMY3 Data Sets", www.nrel.gov/docs/fy08osti/43156.pdf

Download Pdf

IRENA commissioned a study to benchmark the methods for evaluating the technical and economic potentials through 11 major references. The analysis compares the entry parameters considered by each study, and highlights the difference between the approaches.

Wind resource analyses  comparison of the datasets

Wind resource analyses  dimensioning parameters to estimate the technical potentials

1 Suitable areas are reduced by glaciers, sand dunes (both with +10 km security zone) saltpans, hydrological areas, protected areas. 2 The factor depends on competing land use options. 3 Centralised PV systems are defined as semi- to large-scale systems (>10 kWp capacity), installed at the ground in areas with little competing land use options. Decentralised PV systems are defined as small- to medium-scale systems (100 W to 10 kWp) for domestic electricity supply, installed at or close to houses, utilities or industries (see Hoogwijk, 2004). 4 Exclusion parameters: topography, land cover, hydrology, geomorphology and land use. 5 Water bodies, urban areas and military bases were assumed to be not available for development and were excluded from consideration 6 For solar PV technology exclusion parameters: no land area is excluded and no space factor between modules was included. For onshore wind technology land available depends on land use; a suitability factor has been applied for each land use type (excluding tropical forest, agricultural lands and grasslands, urban areas, natural reserves, limited availability for regular forests). 7 For solar CSP technology exclusion parameters: urban area, mountainous areas, natural reserves, agricultural and forestry areas; and assumed that due to socio-geographical reasons, the remaining land area is available at 5%. [Hofmanet al., 2002]. For wind offshore technology exclusion parameters: area competition for other functions such as fishery, oil and gas extraction, natural reserves. 8 Solar power potential could be developed by grid based applications (central generation and building integrated PV) and off-grid applications (solar home systems, health and religious institutions, schools and water pumping) according to the area designated for each application in a country scale. Wind power potential is evaluated by eliminating preserved areas such as natural parks, sanctuaries and wildlife reserves, forest lands and water bodies. 9 According to SRES-scenarios [IMAGE-team, 2001] 10 For solar centralised grid-connected and wind application restrictive parameters are urban areas, the roughness length and suitability factor which takes into account the land-use category based on IMAGE 2.2, mainly excluding natural reserves and tropical forests. For decentralised grid-connected application, roof-top areas are estimated according to the GDP per capita. 11 Screening criteria correspond to criteria that impact the economic and technical feasibility of renewable power production; such as supportive policies, PV favourable electric power regulatory and water access among others. The list of criteria was then evaluated for its ability to be used in GIS mapping, and GIS data availability and sources were discussed. 12 Screening criteria are: transmission access and capacity available, minimum parcel of 40 acres for central generation and slope of land area must be within the acceptable ranges among others. 13 According to CLC nomenclature please refer to p. 4485-4486 of Dominguez Bravo et al., 2007. 14 Parameters such as: resource constraints, geological characteristics, transmission infrastructure, environmental and federal land areas, high slope grades, overall constructability and suitability of the site.

Wind resource analyses  dimensioning parameters to estimate the economic potentials

Wind resource analyses  dimensioning parameters for the technology scenarios

Download Pdf

Authors: Dave Renne (a), Nathalie Ledanois (b), Nicolas Fichaux (c).

Contributing authors: Doug Arent (d), Jake Badger (e), Carsten Hoyer-Klick (f), Sean Whittaker (g).

The authors would like to thank the reviewers for the first version of this report: Andrew Tindal (h), Michael Taylor (c), Nate Blair (d), Thierry Ranchin (j), Veli-PekkaHeiskanen (k), Francis Yamba (k), Maged Mahmoud (l), David VillarFerrenbach (m), EmanueleTaibi (n) Won Jung Lee (c).

Institutions:
• President of ISES, former Senior scientist at the National Renewable Energy Laboratory (NREL),
• Independent consultant,
• International Renewable Energy Agency (IRENA)
• National Renewable Energy Laboratory (NREL)
• Technical University of Denmark, DTU Wind Energy
• German Aerospace Centre (DLR)
• International Finance Corporation (IFC)
• Germanischer Lloyd/Garrad Hassan (GL/GH)
• Southern African Development Community (SADC)
• Mines ParisTech
• Centre for Energy, Environment & Engineering, Zambia (CEEEZ)
• Regional Centre for Renewable Energy and Energy Efficiency (RCREEE)
• ECOWAS Regional Centre for Renewable Energy and Energy Efficiency (ECREEE)
• Secretariat of Pacific Community (SPC)

The designations employed and the presentation of materials herein do not imply the expression of any opinion whatsoever on the part of the Secretariat of the International Renewable Energy Agency concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The term "country" as used in this material also refers, as appropriate, to territories or areas.

Download Pdf