Remote Sensing Applied to Developing Renewable Energy
Energy. It is the fundamental essence of our universe. Even matter, energy’s cosmic complement, can be reduced to quantities of energy via Einstein’s famous equation E=mc2. It is the energy of the sun that allowed for the evolution of life on this planet and sustains life today. The increasing ability to exploit energy resources has defined the progress of human civilization. Indeed, energy is a universal factor within economies and across societies, and thus much of modern governmental policy revolves around securing and controlling energy resources.
Several problems plague the ways that humans currently collect energy resources and use them. The U.S. Department of Energy predicts that as world population approaches 10 billion around 2030, global energy demand will surpass supply by 20 percent. Moreover, the negative externalities of current fossil-based energy sources range from health and environmental consequences to those of international politics and security. For these reasons, there is a consistent motivation toward the use of alternative energy sources. The power of light from the sun, currents of wind, and heat from the earth’s interior have been used in traditional, non-industrial capacities around the world for millennia. Further development of these renewable sources of energy for modern industrial society is one important way of mitigating the problems of decreasing global energy supply and increasing energy demand, and of the negative external effects that accompany generation from fossil fuels.
Satellite-based remote sensing can aid in realizing the potential of renewables. While not directly involved in the generation of energy, application of this complementary technology is in the supply of information for determining the optimal location of generating facilities, as well as for operational decisions of generating facilities and electric power grid management. Increased investment into the research and development of environmental satellites by government is essential in the interest of sound and sustainable policies, both for energy and for space. Three promising forms of renewable energy – solar, wind and geothermal power – can benefit from on-orbit remote sensing.
|Figure 3 - Reflected solar energy image from NASA.|
Much experimental work has been done over the past decade both in Europe and in the United States to use satellite-based imaging technology as a tool in resource assessment for solar energy. Inaccuracy of solar irradiance (sunlight intensity) information can lead to the selection of less than optimal site choices, system sizes, and performance reliability. Government-owned hardware already exists to provide data as a public good for a number of meteorological and environmental applications, among them determining solar irradiance at the surface of the earth. Satellite techniques offer a relatively inexpensive method of assessing solar irradiance over large areas. Even low-resolution data using moderate-level technology could provide a useful starting point for subsequent ground-based measurements, saving both time and money.
Currently, NASA has several experimental satellites yielding data applicable to surface solar irradiance assessment, including Terra, Aqua, and the Tropical Rainfall Measurement Mission (TRMM). These satellites employ instruments such as the Cloud and Earth Radiant Energy System (CERES) and the Moderate Resolution Imaging Spectroradiometer (MODIS).
The most advanced use of satellite data for solar resource assessment has come with the European Commission’s HELIOSAT-3 project, conducted under the Energy, Environment and Sustainable Development program. This project aims to address several problems identified by the solar energy industry concerning solar irradiance data. Specifically, the uncertainty of solar irradiance estimates is too high, the spatial and temporal resolutions are too low, and many variables such as angular and spectral distribution are undetectable even with ground-based equipment. Objectives for HELIOSAT-3 are to supply irradiance data by angular and spectral distribution, to double spatial and temporal resolution, and to reduce errors to less than 5 percent, 10 percent and 20 percent RMSE (Root Mean Squared Error) on a monthly, daily and hourly basis, respectively.
Until now, the accuracy of solar irradiance assessment using satellite data has been limited by a lack of information about the state of the atmosphere (e.g. clouds and aerosols). To improve accuracy, HELIOSAT-3 exploits data from the European Space Agency’s new METEOSAT Second Generation (MSG) satellites. As opposed to the first generation of imagers offering only 3 wide frequency bands measuring data with 8-bit coding, MSG imagers collect light in 12 narrow bands with 10-bit coding. This improvement translates to an increase in spectral and radiometric resolution, which will allow for the identification of atmospheric parameters with much greater detail, and thus will lead to a better estimation of the true conditions on the ground. MSG also improves accuracy in time and space from pictures of 2.5 km resolution every 30 minutes to 1 km resolution every 15 minutes (Heinemann 1999).
|Figure 2 - Sources of light for satellite imagers|
The performance of HELIOSAT-3 will be demonstrated for applications not only in the sighting of solar power generation facilities, but in the performance monitoring of existing grid-connected facilities, in the planning and operation of solar thermal generation, and in the optimization of daylight in buildings. In addition, a key component of this project is to make all data available via an internet-based geographical information system (GIS) so that users in the public sector and in industry can fully utilize the results.
To determine future potential economic value, inexpensive and reliable methods are necessary to compare actual power production with expected power production. Actual values are easily obtained through standard metering, but expected values must be obtained though computer simulation. A key input for these models will be timely, site-specific data, such as that from the PVSAT (photovoltaic satellite) project, an experimental performance checking system, which is funded by the European Commission’s “Energy, Environment and Sustainable Development” program. It uses hourly METEOSAT irradiance data as a cost-effective alternative to installing irradiance sensors and intelligent monitoring devices (Hammer et al. 2000). The experimenters are comfortable with the quality of the satellite data and believe that this gap can soon be filled with progress in the modeling process (Beyer et al. 2001). The next generation of this system, PVSAT-2, is expected to yield 5 percent reductions in costs of system maintenance and power production (http://www.pvsat.com/rahmen.html).
Offshore Wind Vectors
Offshore wind speed is greater and more consistent than wind speed on land. This is the result of the sea breeze effect; convection currents in the air above the shore rise in the morning and in the evening when there is a significant difference between the temperature of the land and that of the sea.
Satellites can be of great utility in yielding information on wind power potential. The importance of accurate wind information is a result of the facts that this power source is uncontrollable, is intermittent and is relatively unpredictable. It is crucial to determine not only the minimum, maximum and average wind power potentials for a given area, but also the variability over different time scales – from the very short-term (hours and days) to the long-term (seasons and years). Areas with variation in wind power potential above 10 percent are not suitable for stable electricity generation.
Satellite-based remote sensing technology with applications for wind energy resource assessment is an idea that has developed only recently. Just as with solar power, there are definite cost-savings advantages to large area measurement of wind speed and direction (wind vectors) rather than the installation, operation and up-keep of extensive site-specific instrumentation.
Because the wind energy resources available over the sea can be greater than those over land, it is in measuring sea winds where satellites can be of substantial benefit to wind energy assessment. Comprehensive, accurate offshore wind vector measurements are essential for the development of generation facilities because energy potential is proportional to the third power of wind speed (Hasanger 2002). Thus, even a small error in measuring speed can translate to a large error in energy potential estimation. Conversely, small amounts of additional speed can translate into additional energy proportional to the third power.
The remote sensing technologies used for determining sea wind vectors are different from those used for solar irradiance, which are all passive; they simply collect the light that is available. In contrast, determining sea winds involves the use of radar, an active system that emits microwaves and measures their reflection, in a technique known as scatterometry. The foundation of this technique rests in the principle that sea surface winds create small, centimeter-sized waves on the surface of the water and these waves respond quickly to changes in wind speed. Sea surface roughness at this scale is thus dependent on instantaneous wind speed.
Scatterometers do not measure sea surface wind speeds directly, but indirectly through the roughness of the water. A transmitter emits radio waves at an angle relative to the sea surface and a receiver measures the backscattered radio waves (see Figure 3). A rough surface will scatter these radio waves such that some will be directed back to the satellite, but a flat surface would reflect the radio waves in a different direction. Backscatter brightness will indicate wind speed, while multiple, simultaneous measurements of the same spot from different angles will indicate direction, thus giving the complete wind vector. Because this is a radar system, it is not dependent on sunlight and not hindered by cloud cover, so measurements can be made day or night, during nearly any weather conditions. The challenges to improving this technology lie in the mathematical and statistical methods used to convert scatterometry data into precise wind vectors and in the calibration of measurements with true ground conditions.
|Figure 4 - Scatterometry measures sea winds indirectly through surface roughness (Kelly 2003).|
The European ERS-2 satellite can perform scatterometry to determine wind vectors using its Active Microwave Instrument; however it can do so only when in “wind mode” and so cannot provide continuous wind data. NASA’s QuickScat satellite, on the other hand, is dedicated to measuring only wind vectors. Like ERS-2 and the Aqua, Terra and TRMM satellites mentioned previously, QuickScat is in a low-altitude orbit, but can scan 90 percent of the Earth’s surface every day. On board QuickScat is the SeaWinds scatterometer – the same instrument that had been carried on the ADEOS II – which is capable of a 25 km spatial resolution and can measure winds between 3 and 20 m/s to within 2 m/s and 20-degree accuracy (see Figure 4). Data are available within 3 hours for weather forecasts and within 2 weeks for scientific research.
Investigations into the value of using satellite radar for wind mapping have been performed in Denmark in anticipation of the construction of a large number of offshore wind farms in the near future. Denmark is home to the first offshore wind farm in Vindeby and the worlds’ largest offshore wind farm in Middlelgrunden, consisting of twenty 2 MW turbines. The Danish government plans to have 4,000 MW of offshore wind capacity available by 2030. One Danish study found that ground-based measurements of wind speed on even a relatively small and flat island in the Danish sea did not represent true conditions on the open water. Using ERS-2 scatterometry data did not yield estimates that were much better; the absolute accuracy of scatterometry-derived wind speeds were an order of magnitude less than what is necessary for wind speed predictions. However, the study concluded that the relative accuracy of this new technology could make it useful and that the results are promising.
As with solar irradiance, further research into satellite-facilitated wind energy assessment is being funded by the European Commission’s Energy, Environment and Sustainable Development program. The goal of the WEMSAR (Wind Energy Mapping Using Synthetic Aperture Radar) project is to demonstrate the abilities of satellite-based radar systems in mapping offshore energy resources, and the project will take advantage of the fact that spatial and temporal resolution of radar systems are constantly improving. Utilizing data from ERS-2, ENVISAT, and the Canadian RADARSAT, WEMSAR should be able to attain wind speed estimates at spatial resolutions of 400 meters (http://www.nersc.no/%7Ewemsar/).
Currently, satellite information is expected only to supplement surface-based data, but this technology is still in the developmental stage. Spatial and temporal resolution of radar instruments will improve, as will the statistical software that converts raw data into useful information.
With respect to investment potential, wind is the fastest-growing source of renewable energy, with generating prices now similar to those of coal-fired plants. However, it is challenging for power grid management authorities to schedule wind power because of its intermittency. In addition, a 1 percent error in the wind speed on the site can lead to almost a 2 percent error in energy forecast, according to Garrad Hassan & Partners, wind energy consultants. Electricity from wind can be more reliable and more profitable with better forecasts of wind speed and direction, which in turn would benefit from more and better satellite data. Advanced modeling techniques for wind enabled traders in the United Kingdom to increase earnings on the Short-Term Power Exchange by 5.8 - 7.5 percent.
Utilities are also trying to gain better wind information to maximize profits. An analysis of the value of wind forecasting to Southern California Edison (SCE) found that the eWind modeling system produced by TrueWind Solutions could reduce power output forecast error by 33 - 50 percent over the course of a year. This means that SCE can avoid many of the imbalances of costs that result from differences in the amount of electricity it predicted to sell and the actual amount it was able to sell, potentially saving $2,000,000 per month, according to the Final Report to the Public Interest Energy Research Program of the California Energy Commission, January 2003. eWind relies on a regular flow of meteorological data from the U.S. National Oceanographic and Atmospheric Administration (NOAA) via electronic file transfer.
In an effort to improve its product, TrueWind Solutions is looking to use additional atmospheric data, especially over the “data sparse” Pacific Ocean, and is considering the possibility of using satellites to collect 3-dimensional, multi-spectral measurements of temperature, moisture and radiance. This will be particularly relevant for offshore generation sites. Remote sensing instrumentation that measures ocean wind speed, such as the SeaWinds scatterometer, will yield very useful information for developing and operating these facilities.
As with solar and wind power, remote sensing can play a key role in the exploration phase of geothermal resource development. These resources are in the form of underground water reservoirs that have been heated because of their proximity to magma close to the earth’s crust. Estimation of the potential from these reservoirs involves information not only on their location, but also on their depth, temperature, pressure and size. In performing nation-wide surveys of potential sites for geothermal power plants, analyses are conducted at different scales. Japan’s New Energy and Industrial Technology Development Organization (NEDO), for example, breaks the task down into three phases. Survey A is conducted at scales of 100 - 300 km2, Survey B utilizes the data from Survey A but at 50 - 70 km2 detail, and finally Survey C is used to study sites at the 5 - 10 km2 scale (“Present Status of Geothermal Energy Development”). Currently, Survey A is performed using temperature-sensing equipment on aircraft, but data from satellite-based remote sensing could be cheaper and more consistent in quality, and could offer more than one method of inferring geothermal activity.
The most intuitive method of exploring for subterranean geothermal reservoirs is to look for areas on the surface that are warm. Measurements of land surface temperature can be performed by high resolution, multi-spectral imagers operating in the thermal infrared region of the spectrum. Many remote sensing satellites have this capability. Infrared channels on LANDSAT 7, for instance, have been used to detect hydrothermal discharge in remote areas of New Mexico and British Columbia (Schulze-Makuch).
NASA’s Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) on board NASA’s Terra satellite is the most advanced imager for this function. Japan’s Institute of Advanced Industrial Science and Technology utilized this high spectral resolution imager in a project to develop geothermal energy on remote islands in the eastern part of Indonesia. Nighttime ASTER images were combined with Synthetic Aperture Radar (SAR) images from the Japanese Aerospace Exploration Agency’s JERS-1 satellite, which provided a digital elevation model of the area and thus a topographical base map. The SAR-derived digital elevation model was also very useful in identifying volcanic surface features. These volcanic features gave researchers an additional set of clues as to the nature and location of potential sites for geothermal wells (Muraoka 2003).
Besides SAR, there are other non-thermal/infrared methods of identifying hidden geothermal resources with remote sensing imagers. For instance, hydrothermal discharges can alter the mineral content of the surrounding land surface. Therefore, the presence of certain types of minerals could offer indirect clues as to the location and characteristics of potential geothermal sites. LANDSAT images have been used by the U.S. Geological Survey to produce mineral classification maps of the Great Basin area in the southwestern U.S. Certain kinds of iron-oxides observed with the LANDSAT data were found to correlate with geothermal systems (Taranik et al. 2004).
Hyperspectral imagers on more modern remote sensing satellites could be even more discriminating among the spectral signatures of different minerals. In addition to the observable chemical changes in the land surface that hydrothermal systems can produce, changes in vegetation can also be observed. Some kinds of chemicals that are associated with hydrothermal discharges are toxic to plants and thus can cause them stress. Symptoms of this stress can include dwarfism and gigantism, both of which are relatively easy to identify in field studies. Other manifestations, however, such as a condition called chlorosis, occur at the cellular level and are difficult to see with the naked eye, but can be detected with high-spectral resolution equipment. Common spectral changes include shifts in the red absorption lines for chlorophyll as well as increases in overall albedo. Researchers at the Energy & Geoscience Institute at the University of Utah and the Geographic Resources Center at the University of Missouri have offered so-called vegetation anomaly mapping techniques as a low-cost method for geothermal exploration (Nash and Hernandez 2001).
As with remote sensing of offshore wind vectors, the utilization of hyperspectral and SAR imaging for geothermal activity is still in its infancy. However, sensor technology has continuously improved and novel techniques for using available data have been created.
The potential economic value of remote sensing of geothermal energy has not yet become a target of study. Right now remote sensing is used for the scanning of large areas, in preparation for narrower, ground-based studies to be conducted at a more detailed level. Satellites can perform Earth observation tasks repeatedly, with greater consistency in data quality, and with lower overall cost than airplane surveys. Satellites also are usually equipped with multiple imaging systems to simultaneously detect many different parameters (thermal, geochemical, botanical, etc.) in concert. Furthermore, satellite-based remote sensing is a successful evolutionary technology, meaning that the expertise of individuals and institutions in creating imaging systems is largely focused on satellite platforms because this is an area of technology that is advancing in general. Improvements in the spatial, temporal, spectral, and radiometric resolution of satellite imagers should be expected to benefit geothermal resource exploration, in addition to solar irradiance and offshore wind vectors.
NASA is currently demonstrating the value of remote sensing data for renewable energy needs in the U.S. A project of NASA’s Earth Science Enterprise called Prediction of World Energy Resources (POWER) has been proposed to coordinate with other government agencies, energy industry associations and individual companies to make better use of NASA meteorology and solar radiation data. The Earth Science Enterprise makes several interesting predictions as to the benefit that POWER services could provide to the United States economy. Using energy demand and cost estimates from the Department of Energy and assuming a growth rate of 25 percent for solar and wind power, ESE calculates the yearly POWER benefits to this part of the renewable energy industry to be $1.5 billion by 2010 and $7.13 billion by 2017 (Whitlock and Stackhouse 2002). This prediction is preliminary and optimistic, but if true this would represent a spectacular value to industry.
The future global energy situation looks increasingly bleak, especially as demand from developing countries increases. Thus renewable energy sources such as solar, wind and geothermal will be very important components in assuring that energy supply can keep up with energy demand. The satellite imaging techniques described here can be useful tools in realizing the full potential of sustainable energy sources.
The largest strides toward developing renewable energy sources are being made in Europe, including the use of remote sensing technologies. In accordance with the Kyoto Protocol, the European Commission has pledged that 22 percent of Europe’s energy will come from renewable sources by 2010. It is therefore not surprising that the European Space Agency announced its plans to fund an international program to use Earth observation services to assist in the development and management of solar, wind and hydroelectric sources (according to the European Space Agency Website, http://www.esa.int/export/esaSA/SEMORAYO4HD_earth_0.html). We have already seen similar efforts from the European Commission with projects such as HELIOSAT and PVSAT.
Other space-faring nations with high demand for energy may therefore wish to explore the possibility of cooperation with Europe in the research and applications of remote sensing technologies to this end. The United States and Japan fall into this category, but emerging space powers such as China and India – having even faster growing energy requirements – certainly do as well.
From rising energy prices to global climate change, the consequences of continued dependence on fossil fuels as a primary energy source will not be borne by one nation alone. Technologies to mitigate these effects should, therefore, be developed through the collaborative efforts of nations with capable space programs.
Stackhouse, P. “Satellites and Sensors: Satellite Missions Impacting the Estimation of Solar Irradiance.” National Renewable Energy Laboratory, 1999.
Heinemann, D. and H.G. Beyer, “Current and Future Data Streams from METEOSAT Geostationary Satellites for Solar Energy Applications.” National Renewable Energy Laboratory, 1999.
Hammer, A. et al. “Surveillance of Photovoltaic Solar Energy Systems Using METEOSAT Derived Irradiances.” EUMETSAT Meteorological Satellite Data Users' Conference, Bologna, 2000, http://www.pvsat.com/rahmen.html
Beyer, H.G. et al. “Accuracy of the Estimation of Monthly Performance Figures of Grid-Connected PV Systems Based On Remote Data Sources.” 17th European Photovoltaic Solar Energy Conference, München, Germany, October 22 – 26, 2001.
Hasanger, F. et al. “On Offshore Wind Energy Mapping Using Satellite SAR.” Canadian Journal of Remote Sensing, Vol. 28, No. 1, 2002, p. 82.
Kelly, P. “Remote Sensing for Wind Power Generation.” Geo Information Science, Department of Geology, Salem State College. http://www.dgl.salemstate.edu/Profs/Young/RS-final/Wind/wind%20scatter.htm
Schulze-Makuch, D. “Exploration of Geothermal Discharge Areas Using Remote Sensing and Hydrological Methods.” University of Texas at El Paso. http://mured.nasaprs.com/report/rt_reports/2001/FAR_UT-ElPaso_SchulzeMakuch_Rpt.pdf
Muraoka, H. “Exploration of Geothermal Resources for Remote Islands of Indonesia.” Institute for Geo-Resources and Environment, AIST Today, No. 7, Winter 2003.
Taranik, J. et al. “Regional Assessment of Exploration Potential for Geothermal Systems in the Great Basin using a Geographic Information System.” Great Basin Center for Geothermal Energy. http://www.unr.edu/geothermal/docs/GeothermalGISProp_Coolbaugh.pdf
Nash, G.D. and M.W. Hernandez “Cost-Effective Vegetation Anomaly Mapping for Geothermal Exploration.” Proceedings of the 26th Workshop on Geothermal Reservoir Engineering, Stanford University, Stanford, California, 29-30 January 2001.
Whitlock, C.H. and Stackhouse, P.W. Jr. “An Estimate of NASA/ESE/POWER Program Benefits to the U.S. from 2002 through 2017.” June 3, 2002. Obtained through Jackson D. Collier, Outreach and Applications, NASA's Earth Science Applications Division.