Satellite Sensing of Geothermal Activity

Satellite Sensing of Geothermal Activity

Using Satellite Images to Identify Anomalous Geothermal Hotspots in Malawi

Finding more efficient and cost effective means to find electrical grade geothermal systems is important for the development of more environmentally friendly generation of electrical power. This is important in Malawi due to its current inability to meet generation demand. The use of satellite systems to detect surface temperature anomalies indicative of areas that deserve further exploration has proven to be an effective means of exploration in the western United States and in other parts of the world. This research will explore the use of satellite spectrometry in the assessment of potential geothermal systems and how this relates to further exploration of geothermal systems in Malawi.

The key question in geothermal exploration is where to drill. Finding efficient means to identify target areas is a key to the success of a geothermal generation project. Satellite data is relatively inexpensive when compared to older exploration methods. Older methods are hit and miss. Satellite data can be used to quickly identify areas that deserve closer exploration through ground verification methods (Jones, 2009, pp. 5). Once a potential site has been located, an exploratory well must be drilled and a flow test performed to assess the potential and quality of the site (Maroake, 2003, pp. 15). Satellite imagery allows for survey of initially larger areas then ground screening methods. At the present time, satellite spectroscopy is used as a screening method to identify targets of interest.

Currently, in the U.S., California is the leader in geothermal power production, but Nevada has tons of hotspots and is expected to rely more on geothermal energy in the past (Tavares, 2009). Geothermal energy is an excellent option for remote areas of the world, such as Eritrea on the Red Sea (Abraha, 2005). This is combined with satellite analysis of ground deformation that is associated with geothermal activity in the area (Fialko & Simons, 2000, pp. 14). These two methods confirmed each other in COSO, California. InSAR observations of ground deformation in the area coincided with satellite thermal imaging.

Geothermal Energy Potential in Malawi

Malawi is located in the southernmost sector of the East African Rift. This area is characterized by rifting, volcanism, crustal thinning and areas of enhanced geothermal gradients. Few volcanoes are still active in the area. However, there are a number of geothermal hotsprings present in the area, making it an area of interest for the generation of geothermal power (Allen, Cawkwell, & Dade et al., 2009, pp. 1-5). A number of surface geothermal spots are located in Malawi, and it is hoped that many more will be found that are not currently apparent on the surface.

Currently, Malawi relies on hydroelectricity for much of its generation. However, environmental degradation has led to a decrease in electrical generation that can result in a decrease of as much as 75%, These decreases have had a negative impact on the industrial sector of the country and the country must now consider searching for other forms of power generation. Their geological location makes geothermal energy a viable alternative to their current crisis (Dulanya, 2006, p. 1). However, methods will have to be developed that will allow for more efficient exploration of their geothermal resources.

All geothermal systems are the result of penetration of meteoric fluids into the crust, causing the subsequent heating and upflow of those liquids (Kratt, Coolbaugh, & Calvin, n.d. pp. 1). A number of hotsprings exist in both the northern and southern portions of Malawi. Many of them contain sulphur compounds. Some are contaminated with human and animal usage. These hot springs range from luke warm to steaming (Dulanya, 2006, p. 2-5). They range from a neutral to basic pH with various potentials for conductivity (Dulanya, 2006, p. 2-5). These hotsprings warrant further analysis for potential usage as power generation sites.

This research will explore the potential for using remote sensing via satellite spectral analysis as a means to locate new locations and to explore existing ones. At present, satellite spectroscopy is being used to locate geothermal hotspots in several areas of the world. However, as a review of current literature on the topic will reveal, analytical methods are not yet precise enough to warrant usage of satellite imagery as a stand alone analytical method for the detection of geothermal hotspots. The literature indicates that several factors affect the ability to remotely sense potential geothermal hotspots. These range from vegetation in the area to atmospheric conditions. This research will result in more effective search techniques using remote satellite sensing to detect potential geothermal sites that are generation quality with as little reliance on ground techniques as possible. It will help to advance knowledge of geothermal hotspots to make it a more reliable detection method for geothermal hotspots in Malawi.

Geomorphology and the Detection of Geothermal Systems.

Geomorphology is the most common means of exploring areas that have the potential for development as sources for geothermal electrical generation. Several algorithms have been developed based on geomorphologic components to help detect areas of interest. However, many variables exist that can limit the efficiency of these methods (Nash & Johnson, 2003, pp. 663-666). Geographic Information Systems (GIS) have traditionally been underused as a means to detect geothermal hotspots. However, improvements in these systems increase their potential to provide an efficient means for exploration (Nash, 1999). GIS systems were found to be superior to CAD generated maps due to their connection to a relational database (Nash & Adams, 2001, pp. 1). The connection with the database allows the researcher to examine the underlying thermal data that is associated with the landform.

Remote satellite sensing was found to be a useful tool in the detection of a geothermal area in Indonesia. However, this method continues to suffer from interference from sun anomalies due to the high sun elevation angle in the tropical regions. It also suffered from the inability to obtain accurate data due to continual clouds due to volcanic feathers in the region (Urai, Muraoka, & Nasution, 2002, pp. 99-108). These same concerns may affect the ability to use this technology in Malawi. Although, Malawi has not currently active volcanoes, it has several hotsprings that produce heavy steam. This may limit the ability of satellite sensory equipment to penetrate the steamy cover. Malawi may also be subject to high sun elevation and the anomalies that are caused by it. This factor will have to be carefully examined in the course of the study.

As geothermal water evaporates, it many leave behind certain mineral deposits, such as borate minerals, whose spectral fingerprint can be used to identify likely sites of geothermal activity (Kratt, Coolbaugh, & Calvin, 2006, pp. 435). The ProSpec TIR VS2 was successfully used to identify a new blind geothermal target in the Columbus Salt March Playa in Nevada. Indicator minerals were the key method used to identify the presence of geothermal activity (Kratt, Coolbaugh, & Peppin et al., 2009, pp. 481). Boron levels were used to detect the presence of thermal springs and wells in the Great Basin region (Coolbaugh, Kratt & Sladek et al., 2006, pp. 395). Mapped imagery taken during the day and night can reveal surface anomalies due to geothermal activity by examining known geothermal resources for the presence of hydrothermal alteration of clays and sulfates (Calvin & Coolbaugh, 2003, p. 1).

Remote mapping of minerals was achieved using the Spatially Enhanced Broadband Array Spectrograph System (SEBASS). Atmospheric conditions had to be subtracted out in order to gain a clear view of the surface radiance using this system (Vaughn, Calvin, & Taranik, 2003, pp. 55). The data was verified using a ground study. Several materials were found to interfere with the accuracy of the readings. These included natural vegetation, soil types, man-made materials, roofs, roads, mine tailings and other materials. However, for the most part, these presented a specific signal and could be differentiated from ground level mineral concentrations.

The presence of surface calcium carbonate deposits is associated with elevated subsurface geothermal reservoir temperatures and the presence of hotsprings on the Pyramid Lake Paiute Reservation. However, those associated with these calcium carbonate deposits are typically not electrical grade (Coolbaugh, Lechler, & Sladek, et al., 2009, pp. 461-463). Remote sensing and hyperspectral imagery with a spectral resolution of 15-20 nm was used to identify areas of interest prior to field reconnaissance work (Coolbaugh, Faulds, & Kratt et al., 2006, pp. 61). However, it may be noted that not all calcium carbonate formations represent active geothermal activity. In the case of TIR data over the Buffalo Valley, these deposits were the result of a dry lake (Littlefield & Calvin, 2009, pp. 498).

The enhancement of oxides and hydroxyls, along with subdued vegetative spectral characteristics was used to successfully analyze hydrothermal activity in Baja California, Sur Mexico (Bancora Prol-Ledesma, 2008, pp. 184). Aside from spectral anomalies, satellite magnetic data was also used to detect heat flux underneath the Antarctic Ice Sheet (Maule, Purucker, & Olsen et al., 2005, pp. 404). These studies suggest that thermal fingerprints of an area can be affected by several factors. Some of these factors enhance the ability to detect geothermal activity in an area, while others impede it. The presence of these factors must be taken into consideration as they relate to the specific area being surveyed. A complete study of these factors and their potential affects on the ability to use remote sensing techniques will have to be explored in order to understand how they affect the use of satellite for remote sensing in Malawi.

The use of thermal infrared (TIR) data from ASTER spaceborne instrumentation was used successfully to detect surface temperature anomalies in the Coso geothermal field in eastern California (Eneva, Coolbaugh, & Bjornstad et al., 2007, pp. 335). One of the key difficulties that was found in the use of TIR data from ASTER is that thermal inertia from different types of vegetative matter can make true geothermal anomalies difficult to identify using spaceborne data alone (Dudley-Murphy & Nash, 2003, p. 645). A study of the thermal inertia of the plant material in Malawi would have to be undertaken so that its affects could be subtracted out of the data in order to detect geothermal anomalies in the area to be surveyed.

Items Affecting Signal Strength and Clarity

As with any spectroscopy technique, a number of elements can affect the ability to detect the desired target. In the laboratory, one has better control over these elements. For instance, one can inject a known gas into the tube, control temperature and take steps to eliminate known contaminants. However, this is more difficult in the natural setting. Spectroscopy techniques from space work in a similar fashion to those in the laboratory except for one important element. Analyses conducted from space do not afford the ability to control the environment in which the analysis is performed. There are a number of factors that can affect the signal and the accuracy of the analysis when it is performed from space. The following will examine these factors and their affect on the ability to perform an accurate assessment.

. Vegetation appears to present one of the key difficulties in anomaly mapping for geothermal exploration. Geobotanical anomalies may accompany mineral deposits and hydrothermal convection systems (Nash & Hernandez, 2001, pp. 1). Therefore, vegetation may not represent unwanted interference all of the time, but may be the clue needed to confirm the anomaly. The affects of vegetation in Malawi will have to be studied to determine the importance of potential interference to signal strength, impedance and phase shift. Colinearity of certain mineral mixes may also have an impact on the spectra. In the case of calcium carbonate, this anomaly caused by colinearity of the complex mixture indicates hydrothermal convection (Nash & Johnson, 2002, pp. 8). In a study conducted in Turkey, mineral identification was impeded by areas of cloudiness as well as dense forests (Dogan, 2008, pp. 224). Geothermal activity may also be responsible for flow patterns in glaciers and ice sheets in Iceland (Bourgeois, Dauteuil, & Vliet-Lanoe. 1999. pp. 74).

Other factors were found to affect the ability to detect geothermal temperature fluctuation anomalies, including elevation and other topographical features, making it difficult to distinguish surface fluctuations from subsurface anomalies (Eneva & Coolbaugh, 2009, pp. 467-470). A comparison of daytime and nighttime images reveals that these comparisons can be an important tool in distinguishing true subsurface geothermal anomalies from surface and atmospheric interference. However, much more work is needed in this area to understand the application of these paired data sets (Eneva, Coolbaugh, & Bjornstad et al., 2007, pp. 1-7).

The pixel in the hyperspectral data set represents a combination of the end-member materials, with those of areal abundance representing the pixel color. However, each pixels distinct and must be analyzed further to determine of indicator minerals for geothermal hotspots are present (Kratt, Calvin & Coolbaugh, 2005, pp. 271-276). This will be an important area for further development of spectral techniques, as many minerals have similar absorption features, as was the case with chalcedony and opal in a recent analysis (Kratt, Calvin, & Coolbaugh, 2004, pp. 1304). Current efforts in research are focusing on the ability to unmix spectra and cancel the affects of atmospheric interference through the use of algorithms (Pal & Nash, 2003, pp. 669-672).

Spectral analysis using wavelengths of 0.4 to 2.5 ?m have been used as the means to identify spectral signatures in the laboratory. These same analysis techniques are now being used in spaceborne methods of spectral analysis (Kratt, Calvin & Lutz, n.d. pp. 1-9). At the present time, these results are being confirmed using handheld GPS and pocket personal computers to provide field verification of results (Coolbaugh, Sladek, & Kratt et al., 2004, pp. 321-324). Spectral analysis of the same set of core samples using both Analytical Spectral Devices (ASD) and Fourier Transform Infrared Spectrometer (FTIR) demonstrated the differences in the two analytical methods (Calvin, Kratt, & Fauldsm 2005, pp. 1-18). Further fieldwork is needed to understand these differences and their affect on data analysis fully.

This section of literature highlights the need for further research in the area of limiting or eliminating disturbances that are due to environmental factors that are beyond the control of the researcher. Currently, researcher is going in the direction of the development of algorithms that correct for many of these factors. However, statistical elimination of these factors has advantages and disadvantages. From one perspective, algorithms can be developed that effectively eliminate the troublesome interference. However, they can also eliminate part of the desirable part of the signal as well. One must do many tests in the area to be assessed in order to confirm the results of statistical elimination of confounding variables in the field. There is still much work to be done in this area regarding remote sensing using satellite imagery. This research hopes to aid research in this direction, with the ultimate goal of making remote sensing by satellite a more reliable means to detect geothermal hotspots in Malawi and in the rest of the world.