USF Libraries
USF Digital Collections

Self-potential anomalies and CO2 flux on active volcanoes

MISSING IMAGE

Material Information

Title:
Self-potential anomalies and CO2 flux on active volcanoes insights from time and spatial series at Masaya, Telica, and Cerro Negro, Nicaragua
Physical Description:
Book
Language:
English
Creator:
Lehto, Heather L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Hydrothermal system
Fluid flow
Diffuse degassing
Mass flow
Continuous monitoring
Dissertations, Academic -- Geology -- Masters -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Considerable effort worldwide has gone into monitoring heat and mass transfer at active volcanoes, as this information may provide clues about changes in volcanic activity and impending eruptions. One method used is the self-potential (SP) method, which has been employed on volcanoes to map hydrothermal systems and structural features and to monitor changes in the hydrothermal system due to volcanic activity. Continuous monitoring of SP has been employed on a few volcanoes and has produced encouraging results. This study presents new time series data collected from continuous monitoring stations at Masaya and Telica, and spatial series data from Masaya, Telica, and Cerro Negro, three active volcanoes in Nicaragua. The primary goals of this study were to determine whether correlations between SP anomalies and CO2 flux exist and to investigate temporal variations in temperature, SP, rainfall, and barometric pressure.To achieve these goals, SP and CO2 flux surveys were conducted on Masaya, Telica, and Cerro Negro, and continuous monitoring stations were installed on Masaya and Telica. The continuous monitoring station on Masaya recorded temperature, SP, rainfall, and barometric pressure. The station on Telica recorded temperature and SP. Profiles collected on Masaya and Cerro Negro show broad correlation between SP and CO2 flux. However, profiles on Telica revealed virtually no SP anomaly or CO2 flux for the majority of the profile, at the time of data collection. Data collected from the continuous monitoring station at Masaya showed a persistent positive SP anomaly that fluctuated between 60 and 240 mV. Rainfall was seen to supress the anomaly for time scales of several hours to several days. Correlations between temperature, SP, and barometric pressure were also seen at Masaya.Curiously, no increases in SP were seen during two temperature transients that occurred during volcanic activity in June and October. Continuous monitoring data from Telica showed only decreases in temperature and SP, which coincided with rainfall. The continuous monitoring data collected in this study and others have begun to provide a better understanding of the nature of SP anomalies, which may aid in the development of the SP method as a volcano monitoring tool.
Thesis:
Thesis (M.S.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Heather L. Lehto.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 91 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001920230
oclc - 187972868
usfldc doi - E14-SFE0002108
usfldc handle - e14.2108
System ID:
SFS0026426:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Self-Potential Anomalies and CO 2 Flux on Active Volcanoes: Insights from Time and Spatial Series at Masaya, Teli ca, and Cerro Negro, Nicaragua by Heather L. Lehto A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology College of Arts and Sciences University of South Florida Major Professor: Charles Connor, Ph.D. Diana Roman, Ph.D. Mark Stewart, Ph.D. Sarah Kruse, Ph.D. Date of Approval: July 10, 2007 Keywords: hydrothermal system, fluid flow diffuse degassing, mass flow, continuous monitoring Copyright 2007, Heather L. Lehto

PAGE 2

Acknowledgments First and foremost I would like to thank Armando Saballos, Allan Morales, Wilfried Strauch, and the rest of the staff at the Instituto Nicaragense de Estudios Territoriales (INETER) for their support a nd collaboration on this project including providing vehicles, support staff, and equipmen t. Their help has been invaluable. I would also like to thank Ward Sanford for his support and help throughout the course of this project and during some challe nging field work. I would also like to thank Sophie Pearson for her help in the field and for the many lengthy discussions we had about the data. At last but certainly not least, I would like to thank my committee members: Chuck Connor, Diana Roman, Mark St ewart, and Sarah Kruse for their much needed wisdom and guidance.

PAGE 3

i Table of Contents List of Figures ii Abstract v Introduction 1 Origins of Self-Potential (SP) Anomalies 3 Thermoelectric Coupling 4 Electrochemical Effect 5 Oxidation-Reduction Reactions 5 Electrokinetic Coupling 6 Rapid Fluid Disruption (RFD) 8 SP surveys on active volcanoes 8 Spatial Variation 9 Temporal Variation 10 Geologic Setting 12 Methods 16 SP and CO 2 Flux Profiles 16 Continuous SP Monitoring 17 Design and Performance of Pb-PbCl 2 Electrodes 20 Results 22 SP and CO 2 Flux Profiles 22 Continuous SP Monitoring: Masaya Volcano 24 Continuous SP Monitoring: Telica Volcano 37 Discussion 39 SP and CO 2 flux profiles 39 Continuous SP monitoring: Masaya Volcano 40 Continuous SP monitoring: Telica Volcano 47 Conclusion 48 References 50 Appendices 56 Appendix 1: Additional Plots, Masaya 57 Appendix 2: Additional Plots, Telica 84

PAGE 4

List of Figures Figure 1. Schematics of electrical potenti al generation and the relation to the polarity of the SP anomaly. 2 Figure 2. Diagram of the charge distri bution at the mineral/fluid interface and the zeta potential across the shearing plane, S. 7 Figure 3. Location map and shaded relief map of the study area. 13 Figure 4. Shaded relief maps of Masaya showing locations of the Comalito Site, Fumarole Fiel d Site, and Hilltop Site. 14 Figure 5. Left: Shaded relief maps of Te lica showing site location. Right: Shaded relief maps of Cerro Negro showing site location. 15 Figure 6. Photographs of sites at Masaya. 18 Figure 7. Photographs of sites at Telica and Cerro Negro. 19 Figure 8. Left: Schematic design of Pb-PbCl 2 electrodes from Petiau (2000). Right: Photograph of electrodes c onstructed using design by Petiau (2000). 21 Figure 9. Graph of self-potential and ra infall showing noise added when new electrodes were installed on Masaya. 21 Figure 10. Top: SP and CO 2 Flux profiles from Masaya; Comalito Site, Line 1. Bottom: SP and CO 2 Flux profiles from Masaya; Hilltop Site. 22 Figure 11. Top: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 0. Center: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 1. Bottom: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 2. 23 Figure 12. Top: SP and CO 2 Flux profiles from Telica. Bottom: SP and CO 2 Flux profiles from Cerro Negro. 24 ii

PAGE 5

iii Figure 13. Temperature and rainfall da ta (top) and SP and rainfall data (bottom) for the period from May 27, 2006 through March 19, 2007 from Masaya. 25 Figure 14. Spectrograms of SP (above) a nd temperature in the thermocouple at 90 cm (below). 26 Figure 15. Rainfall magnitude versus SP recorded during times of heaviest rainfall in July and October. 26 Figure 16. Temperature and rainfall da ta (top) and SP and rainfall data (bottom) recorded during the te mperature transient in June 2006 at Masaya 29 Figure 17. Temperature and rainfall da ta (top) and SP and rainfall data (bottom) recorded during the te mperature transient in October 2006 at Masaya. 30 Figure 18. Spectral estimates of SP (top left), barometr ic pressure (top right), temperature at the thermocouple at 90 cm (bottom left), and rainfall (bottom right). 31 Figure 19. Spectral estimates of SP (top left), barometr ic pressure (top right), and temperature at the thermocouple at 90 cm (bottom) during the June 2006 temperature transient. 32 Figure 20. Spectral estimates of SP (top left), barometr ic pressure (top right), and temperature at the thermocouple at 90 cm (bottom) during the October 2006 temperature transient. 33 Figure 21. Temperature at the thermoc ouple at 90cm, barometric pressure, and SP recorded from February 1, 2007 through February 6, 2007 at Masaya. 34 Figure 22. Temperature at the thermoc ouple at 90cm, barometric pressure, and SP recorded from February 8, 2007 through February 13, 2007 at Masaya. 35 Figure 23. Temperature at the thermoc ouple at 90cm, barometric pressure, and SP recorded from February 15, 2007 through February 20, 2007 at Masaya. 36 Figure 24. Temperature and rainfall da ta (top) and SP and rainfall data (bottom) for the time period from June 1, 2006 through March 21, 2007 at Telica. 37

PAGE 6

iv Figure 25. Spectral estimate of SP and temperature at T1 at Telica, showing dominant 24 and 12 hour periods in SP and 24 and 11 hour periods in temperature. 38 Figure 26. Schematic of the background conditions at Masaya. 42 Figure 27. Schematic of the system at Ma saya during times of heavy rainfall. 44 Figure 28. Schematic of the system at Masaya during times of increased barometric pressure. 45 Figure 29. Schematic of the system at Masaya during times of decreased barometric pressure. 46

PAGE 7

v Self-Potential Anomalies and CO 2 Flux on Active Volcanoe s: Insights from Time and Spatial Series at Masaya, Te lica, and Cerro Negro, Nicaragua Heather L. Lehto ABSTRACT Considerable effort worldwide has gone into monitoring heat and mass transfer at active volcanoes, as this information may provide clues about changes in volcanic activity and impending eruptions. One method used is the self-pot ential (SP) method, which has been employed on volcanoes to ma p hydrothermal systems and structural features and to monitor changes in the hydr othermal system due to volcanic activity. Continuous monitoring of SP has been empl oyed on a few volcanoes and has produced encouraging results. This study presents new time series data colle cted from continuous monitoring stations at Masaya and Telica, a nd spatial series data from Masaya, Telica, and Cerro Negro, three active volcanoes in Ni caragua. The primary goals of this study were to determine whether correl ations between SP anomalies and CO 2 flux exist and to investigate temporal variations in temperatur e, SP, rainfall, and barometric pressure. To achieve these goals, SP and CO 2 flux surveys were conducted on Masaya, Telica, and Cerro Negro, and continuous mon itoring stations were installed on Masaya and Telica. The continuous monitoring stat ion on Masaya recorded temperature, SP, rainfall, and barometric pressure. The sta tion on Telica recorded temperature and SP.

PAGE 8

Profiles collected on Masaya and Cerro Ne gro show broad correlation between SP and CO 2 flux. However, profiles on Telica revealed virtually no SP anomaly or CO 2 flux for the majority of the profile, at the time of data collection. Data collected from the continuous monitoring station at Masaya show ed a persistent positive SP anomaly that fluctuated between 60 and 240 mV. Rainfall wa s seen to supress the anomaly for time scales of several hours to several days. Correlations between temperature, SP, and barometric pressure were also seen at Masa ya. Curiously, no increa ses in SP were seen during two temperature transients that occu rred during volcanic activity in June and October. Continuous monitoring data from Telica showed only decreases in temperature and SP, which coincided with rainfall. The continuous monitoring data collected in this study and others have be gun to provide a better understanding of the nature of SP anomalies, which may aid in the development of the SP method as a volcano monitoring tool. vi

PAGE 9

Introduction Studies of heat and mass transfer have ga ined attention over the past few years in the hope that the information these studies pr ovide can help predic t changes in volcanic activity and impending eruptions. One method us ed to study these phenomena is the selfpotential (SP) method. The SP method has traditionally been used in mining to locate metallic ore but has more recently been applied to geothermal and volcanic areas with promising results (e.g., Hashimoto and Tanaka, 1 995; Zlotnicki et al., 2001; Friedel et al., 2004). The SP method involves the measurement of the electric potential between two points on the ground surface using two non-polarizing electrodes and a high-impedance voltmeter. The electric potential measured by the electrodes is cr eated by the primary flow of ions (heat or fluid flow in the subs urface) which induces a current that interacts with the resitivity structure and creates an electric potential according to Ohms law (Nourbehecht, 1963; Sill, 1983; Revil et al., 2003). An SP survey is conducted by placing one electrode, the reference electrode, in the ground, preferably in an area of no a nomalous flow, and then moving the roving electrode along the survey line or throughout a map area and recording the electric potential between the two electrodes. By c onvention, the roving electr ode is attached to the positive terminal of the voltmeter and th e reference electrode is connected to the 1

PAGE 10

negative terminal. Anomalies are identified in SP surveys when excess positive charge accumulates at the roving electrode, yiel ding a positive SP anomaly, or when excess negative charge accumulates at the roving elec trode, yielding a negative anomaly (Figure 1). Roving Electrode Roving Electrode Reference Electrode Reference Electrode Figure 1. Schematics of electri cal potential generation and th e relation to the polarity of the SP anomaly. Electrical pot ential generation is due to th e interaction of a current, induced by subsurface flow of heat or fluids, with the resistivity structure, according to Ohms law. Left: A positive SP anomaly would be r ecorded due to the conventional electrode setup, for example, where hydrot hermal upwelling occurs near the positive electrode. Right: A negative SP anomaly would be record ed if there is excess negative charge near the positive elec trode, such as where downward flow of fluids occurs. The SP method has been used on volcanoes to detect SP anomalies with a wide range of amplitudes of positive and negative polarity. The spatial distribution of these anomalies has been used to determine the boundaries of structural features, such as fracture zones, as well as hydrothermal syst ems (Massanet and Pham, 1985; Jackson and Kauahikaua, 1987; Michel and Zl otnicki, 1998; Aubert et al., 2000; Finizola et al., 2002; Finizola et al., 2004). Repeated SP surveys have also been used to study changes in the SP anomaly over time scales of several days to years (Dzurisin et al., 1980; Di Maio and 2

PAGE 11

Patella, 1994; Hashimoto and Tanaka, 1995; Zlotnicki et al., 2003; Aizawa, 2004). In addition, a few researchers have recently employed continuous SP monitoring to study the variation of SP on short time scales, with promising results (Hashimoto and Tanaka, 1995; Zlotnicki et al., 2001; Friedel et al., 2004). The purposes of this study are two-fold; first to determine whether SP anomalies detected on Masaya, Telica, and Cerro Negro correlate with CO 2 flux. Second, to determine, through continuous monitoring of te mperature, SP, barometric pressure, and rainfall at Masaya and Telica, whether the SP anomaly persisted throughout the year, what effect rainfall had, what would happen to the SP anomaly during the dry season, and how changes in ground temperature, due to ch anges in volcanic activity, would affect the SP anomaly. Origins of Self-Potential (SP) Anomalies Despite the many SP surveys conduc ted on volcanoes around the world, the source mechanisms that generate these anoma lies are still not comple tely understood. All possible source mechanisms can be underst ood through the theory of irreversible thermodynamics which applies classical th ermodynamic principles to states of nonequilibrium, such as systems in flow (N ourbehecht, 1963). The general equation for coupled flow is written as j j ij iXLJ ( i = 1, 2, 3 n and j = 1, 2, 3 m) (1) 3

PAGE 12

where flow J i (such as heat, electrical current, etc.) is due to a driving force X j (such as temperature gradient, pressure gradient, etc.) and a cross-coupling coefficient L ij and i and j represent the different types of sec ondary and primary flow, respectively. Thermoelectric Coupling A temperature gradient across a sample of rock will produce an electric potential across the rock. This is calle d the thermoelectric effect. Th e driving forces responsible for the heat flow J T (J m -2 s -1 ) and electric current density I T (A m -2 ) are the temperature gradient T (K m -1 ) and electric potential gradient (V m -1 ), respectively. The equations for J T and I T are T JT and (2) T IT (3) where and are the electrical conductivity (S m -1 ) and thermal conductivity (m kg s -3 K -1 ), respectively and and are the Peltier coefficient (kg m 2 s -3 A -1 ) and thermoelectric coefficient (A m -1 K -1 ), respectively (Zlotnicki and Nishida, 2003). The thermoelectric coupling coe fficient C is defined as T / Modeling by Nourbehecht (1963) and Corw in and Hoover (1979) showed that in the case of a buried, spherical body subjected to an elevated temperature (100 C) and centered between two horizonta l layers with thermoelect ric couple coefficients C 1 and C 2 the maximum surface voltage expected is 0.15 (C 1 C 2 ) T Using a realistic value of 0.2 mV/C for (C 1 C 2 ), the maximum voltage expected in this idealized case due to the thermoelectric effect is 3 mV. 4

PAGE 13

Electrochemical Effect An electrical potential can also be ge nerated by chemical diffusion of ions between chemical concentration cells. In this case the driving forces are the concentration gradient (m C -4 mol) and the electric potential gradient (V m -1 ). The chemical flux J C,m (mol m -2 s -1 ) and the electric current density I C (A m -2 ) are expressed as )/ (, mm mmm mCCDTRFCZDJ and (4) ,} )/ ({2 mmm mmm CCZDTRFCZDFI (5) where D m Z m and C m are the diffusion coefficient (m 2 s -1 ), charge number, and concentration of the m th ion (m -3 mol), respectively. F, R, and T are Faradays constant (A s mol -1 ), the gas constant (J K -1 mol -1 ), and absolute temperature (K), respectively (Zlotnicki and Nishida, 2003). Nourbehecht (1963) estimate d that potentials created by electrochemical effects coul d not exceed 20 mV when m odeled using experimentallyobtained values of electrochemical coupling coefficients for various rock samples. Oxidation-Reduction Reactions Volcanic gases containing H 2 S, SO 2 and CO 2 that interact with groundwater can generate negative SP anomalies throu gh chemical reactions that produce , and ions (Massenet and Pham, 1989; Zlotnicki and Nishida, 2003). However, it has been shown that the amplitude of an SP anomaly produced through the common 4SO 3HCO 3CO 5

PAGE 14

oxidation-reduction reactions is not large enough to account for the anomalies observed on volcanoes (Zlotnicki and Nishida, 2003). Electrokinetic Coupling The flow of a fluid through a porous me dium will generate an electric potential gradient along the flow path due to the inte raction of the fluid w ith the electrical double layer. The resulting potential is called the el ectrokinetic, or stream ing, potential. When a fluid comes in contact with certain minerals an electrical double layer is formed along the surface of the mineral. The electrical double layer contains an outer Gouy-diffuse layer and an inner Stern layer. The mineral has a net negative charge, while the outer diffuse layer has a net positive charge. Once a pressu re gradient is established the induced fluid flow shears off the outer diffuse layer along a shearing plane (Figure 2). The electric potential across the shearing pl ane is called the zeta potential and plays an integral role in the electrokinetic or streami ng potential, as the magnitude of controls the charge difference the develops across the shearing plane. The fluid flux J E (m s -1 ) and the electric current density I E (A m -2 ) are induced by the elect rical potential gradient (V m -1 ) and the pressure gradient P (Pa m -1 ), respectively. The equations for fluid flux J E and current density I E have the form ),( gP k LJE (6) ),( gPL IE and (7) 6

PAGE 15

,/0 L P CEI (8) where C is the electrokinetic coupling coefficient (V Pa -1 ), L (m 2 V -1 s -1 ) represents both the electrokinetic coupling coefficient and th e electroosmotic coupling coefficient, k is the permeability of the medium (m 2 ), is the mass density of the fluid (kg m -3 ), is the electrical co nductivity (S m -1 ), is the dielectri c constant (F m -1 ), is the dynamic shear viscosity of the fluid (Pa s -1 ), and is the zeta potential (V). It has been shown that is negative for rocks and minerals with pH above two but becomes positive for pH below two (Ishido and Mitzutani, 1981; Hase et al., 2003). Closer inspection of Equati on (8) shows that a negative will produce a positive SP anomaly while a positive will produce a negative SP anomaly. Figure 2. Diagram of the charge distribution at the mineral/fluid interface and the zeta potential across the shearing plane, S (aft er Antraygues and Aubert, 1993). 0 is the surface potential. In this case is negative, the Gouy-diffuse layer is enriched in positive ions, and a positive SP anomaly results. 7

PAGE 16

Recent modeling by Bolve et al. (2007) of SP anomalies due to groundwater flow suggests that the streaming current density may be directly re lated to the seepage velocity of the groundwater and to the exce ss electrical charge per unit pore volume in the porous material. The authors argue that th e new formulas are better suited for use in SP modeling because they take into account the permeability of the porous medium and can be applied to unsaturated conditions. Rapid Fluid Disruption (RFD) Recently Johnston et al. (2001) propos ed rapid fluid disruption (RFD) as a mechanism for generating SP anomalies on volcanoes. RFD is a process by which charge separation occurs due to rapid vaporization of flui ds and has been proposed to explain high amplitude SP anomalies obser ved on volcanoes in areas far above the groundwater table. Although RFD most lik ely does play a role in generating SP anomalies in some areas, its significance is still hotly debated (Revil, 2002; Johnston et al., 2002; Lewicki et al., 2003). SP Surveys on Active Volcanoes SP surveys conducted on active volcanoes can be put into three basic categories: those that focus only on spatial variations, th ose that include both spatial and temporal variations, and those that c oncentrate on temporal variati ons only. Spatially varying surveys have generally been geared toward s describing the structure of hydrothermal systems on active volcanoes (Jackson and Kaua hikaua, 1987; Aubert et al., 2000; Kanda 8

PAGE 17

and Mori, 2002; Finizola et al ., 2002; Finizola et al., 2003; Fi nizola et al., 2004), or used to demonstrate correlations with gas flow (L ewicki et al., 2003). Some workers have begun collecting repeated or continuous SP maps and profiles to study spatial and temporal variations in the hydr othermal system due to volcanic activity (Dzurisin et al., 1980; Massenet and Pham, 1985; Di Maio a nd Patella, 1994; Hashimoto and Tanaka, 1995; Michel and Zlotnicki, 1998; Sasai et al., 2002; Zlotnicki et al., 2003; Aizawa, 2004). The use of continuous SP monitoring to study temporal variations has also recently been employed on Unzen (Hashimoto and Tanaka, 1995), Merapi (Friedel et al., 2004), and Piton de la Fourna ise (Zlotnicki et al., 2001). Spatial variation Spatial SP surveys have been performed on several volcanoes with mixed results. Although the method is the same, the scales of spatial surveys are very different, which may play a role in the variation in the resu lts obtained. Spatial surveys completed using a coarse grid (50-110 m grid spacing) on volcanoes such as Kilauea (Dzurisin, 1980; Jackson and Kauahikaua, 1987), Misti (Finiz ola et al., 2004), Satsuma-iwojima (Kanda and Mori, 2002), Piton de la Fournaise (Michel a nd Zlotnicki, 1998), Etna (Di Maio and Patella, 1994), Unzen (Hashimoto and Tana ka, 1995), and Fuji (Ai zawa, 2004) all show SP highs centered on the crater or recently active fracture zones and SP lows along the flanks. However, the results of surveys c onducted using a much finer grid spacing the can be quite different. An SP map created on Stromboli, using 20 m grid spacing, shows a complex pattern of SP highs and lows along the flanks and at the summit (Finizola et 9

PAGE 18

al., 2002). The pattern of the highs and lows is also of both shortand long-wavelength, suggesting that SP signals may in fact be more complex than those seen with larger grid spacing. A single profile collected on Merapi volcano in Indonesi a also contains a repeating pattern of SP highs followed by SP lows (Aubert et al., 2000). With a much finer grid spacing of 1 m, even more detail in the SP pattern on volcanoes can be seen. SP profiles collected across an inferred fract ure zone at Masaya (Lewicki et al., 2003) using 1 m grid spacing show a pattern of SP highs followed by SP lows similar to that seen at Merapi. In 2003 a much finer scale (1 m grid spacing) survey was conducted in the summit crater at Stromboli (Finizola et al., 2003). This survey revealed a complex pattern of shortand long-wavelength anomalies. This suggests that SP anomalies on volcanoes are characterized by a much more complex pattern of longand shortwavelength anomalies than previously believed, or mapped. Temporal Variation The use of repeated SP surveys to stu dy temporal changes in the hydrothermal system due to volcanic activity has been employed at several volcanoes with varying results. Repeated surveys on Kilauea (Dzuri sin, 1980), Etna (Di Maio and Patella, 1994), and Piton de La Fournaise (Michel and Zlot nicki, 1998; Zlotnicki et al., 2001) showed increases in SP during times of increased vol canic activity. Changes in SP spanned a wide range of values from 20 mV (Kilauea) to 1500 mV (Piton de La Fournaise). A repeated survey conducted on Fuji (Aizawa, 2004) showed a decrease in SP of 350 mV around the same time as the appearance of a new group of fumaroles. On Miyake-jima 10

PAGE 19

repeated SP monitoring revealed changes in SP leading up to the July 8, 2000 crater collapse of +10 to +25 mV when both electrodes were located outside of the future crater boundary and -10 to -25 mV when one electrode was positioned inside the future crater boundary (Zlotnicki et al., 2003). The grid sp acing of the repeated surveys discussed above was 61 m, 110 m, 50 m, 200 m, and 100 m for Kilauea, Etna, Piton de La Fournaise, Fuji, and Miyake-jima, respectively. Continuous SP monitoring has only been employed on a handful of volcanoes worldwide. Continuous SP monitoring on Un zen during a lava dome extrusion in May 1991 (Hashimoto and Tanaka, 1995) showed the first promising results. The SP station on Unzen consisted of pairs of electrodes spaced 725-1000 m apart with a sampling interval of 40 minutes. Beginning in March 1991 the SP signal at se veral stations began a steady increase until the time of the eruption onset when the signal then leveled off. The maximum increase in SP was 600 mV at the center of the anomaly. At Piton de La Fournaise (Zlotnicki et al., 2003) continuous SP monitoring over several years, using electrodes with 200 m grid spacing, has shown a long term seasonal variation that does not display a uniform change in amplitude. Continuous monitoring around the time of the March 9, 1998, eruption displayed very low frequencies during fluctuations in SP that were postulated to have been in the ULF (ultra-low frequency, 0.1-10 Hz) band, but could not be resolved below the 20 second sampling interval employed. These low frequencies were not seen during the eruption, however; inst ead an increase in SP of 250 mV was seen approximately one day before the eruption started and decreases in SP of 750 mV were detected during times when fr actures were known to have opened. The 11

PAGE 20

increase in SP recorded before the eruption was interpreted as an upward flow of positive ions due to increased volcanic activity. Conversely, the decrease in SP was considered to be due to fluid flow into the recently opened fractures. Continuous monitoring of rainfall, temperature, and SP on Merapi (Fried el et al., 2003) utilized a 20 to 75 m grid spacing and a sampling interval of 20 s for SP and 0.25 s for temperature. Long term correlations were seen between temperature change and SP as well as short-term responses to rainfall in both temperature a nd SP. The short-term responses to rainfall generally consisted of decreases in temperature seen 4 5 h after a rainfall event and increases in SP with maximums seen 2 4 h after a rainfall event. Periodic variations of 12 h were also observed after times of rainfa ll during the dry season and were attributed to changes in atmospheric pressure. The continuous SP monitoring studies conducted so far have begun to highlight some of the issu es that face the use of continuous SP as a monitoring tool, such as grid spacing, sa mpling interval, and phenomena that influence the SP signal (i.e., rainfall a nd periodic variations). Geologic Setting Located along the Central American Vo lcanic Front, Masaya, Telica, and Cerro Negro are three of Nicaraguas most active volcanoes (Figure 3). Masaya (Figure 4), is located 25 km south of the capital city of Ma nagua and consists of a series of basaltic shield volcanoes, calderas, cinder cones, a nd pit craters formed by a series of Plinian basaltic eruptions (Williams-Jones et al., 2003, MacNeil et al., in-press). The two main cinder cones, Masaya and Nindir, house the summit pit craters Masaya, Nindir, and the 12

PAGE 21

currently active Santiago crater The first historical erupti on, described by the Spanish in 1524, occurred at Nindir crater. Activity at Nindir crater consisted mostly of lava lake formation, the crater became inactive wh en Santiago crater was formed around 1859. Activity at Santiago crater consists mostly of periods of passive degassing during which times incandescence or a lava lake is sometime s visible in the crater vents. The most recent activity has been confined to reports of incandescence in Santiago crater from June through October 2006 and January through Marc h 2007, a vent widening event in June 2006, and the establishment of a new ve nt and lava lake in October 2006. Figure 3. Location map and shaded relief map of the study area. 13

PAGE 22

Figure 4. Shaded relief maps of Masaya showing locations of the Comalito Site, Fumarole Field Site, and Hilltop Site. Telica (Figure 5), located approximately 110 km northwest of Managua in the district of Leon, is a 1061 m high basaltic a ndesitic stratovolcano. The volcano consists of several pit craters along an east-west ali gnment. The active crater is approximately 400 m wide and 200 m deep with vertical or overhanging walls and a 40 m wide vent (Roche et al., 2001). Current activity at Te lica consists of gas, steam, and minor ash eruptions on August 4 and 6, 2006, December 11 and 27, 2006, January 9, 2007, and February 6, 15, and 17, 2007. Cerro Negro (Figure 5) is a 250 m hi gh basaltic cinder cone located 70 km northwest of Managua. The cone was formed in 1850 and has erupted frequently with its most recent eruption in 1999. Eruptions at Cerro Negro have ranged from Strombolian to sub-Plinian and commonly pr oduce flank vents (LaFemina et al., 2004). Cerro Negro 14

PAGE 23

contains two active cones: a main cone and a smaller cone which formed inside the main cone during the 1995 eruption. Lava fl ows from the 1995 eruption breached the northeast rim of the main cone. Activity dur ing the 1999 eruption was mostly confined to flank eruptions with no activity inside the main crater (LaFemina et al., 2004). Figure 5. Left: Shaded relief maps of Telica showing site location. Right: Shaded relief maps of Cerro Negro showing site location. 15

PAGE 24

Methods SP and CO 2 Flux Profiles SP and CO 2 flux profiles were collected on Ma saya in May 2006, Telica in June 2006, and Cerro Negro in August 2006, using 1 m grid spacing for SP measurements and 2 m grid spacing for CO 2 flux measurements. The profiles ranged in length from 20 m to 100 m, depending upon anomaly width. SP measurements were made using nonpolarizing Pb-PbCl 2 electrodes available commercially. CO 2 flux measurements were made using a LICOR Li-800 portable gas fluxmet er. Profiles on Masa ya were located at the base of Comalito cinder cone (Comalito Site Figures 4 and 6), at a site located on the flank of the volcano above the Comalito Site (Hilltop Site, Figures 4 and 6), and at a fumarole field located between the Hilltop Site and the Comalito Site (Fumarole Field Site, Figures 4 and 6). The profiles on Telica were located along the crater rim, next to the seismic station (Figures 5 and 7). The profiles on Cerro Negro were located along the southwest rim of the main crat er (Figures 5 and 7). All su rveys were completed in areas with minimal elevation change so as to a void including any topogra phic effects in the SP data. 16

PAGE 25

Continuous SP Monitoring Continuous SP, temperature, and rainfa ll measurements were recorded at the Comalito Site on Masaya (Figures 4 and 6) from May 2006 through March 2007. SP and temperature measurements were recorded next to the seismic station at Telica (Figures 5 and 7) during the same time period as at Masaya. Continuous data we re collected at five minute intervals using a Cambell Scien tific datalogger, non-polarizing Pb-PbCl 2 electrodes, and chromel-alumel thermocouples. Two sets of electrodes were used for the continuous SP readings. The first set wa s used from May through August 2006 and was purchased commercially. The second set of electrodes was constructed using a design by Petiau (2000) and was used between August 2006 and March 2007. The construction and performance of these electrodes will be discussed in the following section. The continuous SP monitoring station lo cated at Masaya was placed inside a metal cage located on a fumarole (Figure 6). The datalogger, solar panel, rain gauge, four thermocouples, and one electrode were in stalled inside the cage. One electrode and one thermocouple were installed outside the cage. The electrodes were placed ~6 m apart, one at the top of the slope inside the metal cage and one at the bottom of the slope (Figure 6). Both electrodes were placed a few cm into the ground and secured so as to maintain contact with the ground. Four of the five thermocouples were buried at depths of 33 cm, 65 cm, 90 cm, and 150 cm inside th e cage, and one thermocouple was installed in a fumarole ~7 m away from the others. 17

PAGE 26

Line 0 Line 2 Line 1 Electrode 1 Electrode 2 Figure 6. Photographs of sites at Masaya. Top Left: Comalito Site. Datalogger, solar panel, and rain gauge visible inside the cag e. One electrode and five thermocouples located inside the cage. The other electrode is located down the slope to the right side of the photo and one thermocouple located in a fu marole to the left side of the photo. Approximate location of SP and CO 2 flux profiles are shown by black lines. Top Right: Comalito Site. Approximate location of electrodes used for continuous SP monitoring shown. Photo courtesy of D. Roman, 2007. Bottom Left: Fumarole Field Site. Approximate location of SP and CO 2 flux profile lines are shown as black lines. Bottom Right: Hilltop site. Datalogger is shown in cente r-left of photo. Approximate location of SP and CO 2 flux profiles are shown as black lines, arrow indicates prof iles continue off edge of photo. 18

PAGE 27

Electrode 1 and T1 Electrode 2 and T3 T2 Figure 7. Photographs of site s at Telica and Cerro Negro. Left: Seismic Station at Telica. Datalogger is located inside seismic station. El ectrodes and thermocouples are located down slope at the right side of the photo. SP and CO 2 flux profiles taken along rim next to seismic station. Right: Location of SP and CO 2 flux profiles at Cerro Negro. Profiles taken along the measuring tape shown. At Telica the continuous SP monitoring station included a datalogger, three thermocouples, and two electrodes. The datalogger was installed inside a seismic station maintained by staff at INETER and located next to a slope containing a bank of fumaroles. The thermocouples were buried at three levels down the slope; the one farthest downhill (T1) was placed directly in the fumarole, one (T3) was placed at the top of the hill in an inactive ar ea, and the third (T2) was pl aced roughly halfway between the other two in the fumarole. On e of the electrodes was installe d next to T1 and the other installed next to T3 (Figure 7). This stati on did not include a rain gauge: however, rain data collected from a gauge located ~26 km to the northeast and averaged over a day was supplied by INETER and is included in Figure 13. 19

PAGE 28

Design and Performance of Pb-PbCl 2 Electrodes In August 2006 the commercial Pb-PbCl 2 electrodes used for continuous monitoring at Masaya and Telica were replaced with Pb-PbCl 2 electrodes constructed after a design by Petiau (2000) as the porous ceramic bottoms of the commercial set of electrodes had begun to dissolve (Figure 8). The electrodes were constructed by mixing kaolinite, PbCl 2 and KCl to form a paste which filled a PVC pipe with a porous ceramic bottom. A lead wire was soldered to a copper wire and secured to a PVC cap, which was then used to seal the elec trode. The cost to produce one electrode was approximately $20, while the cost of one commercial electrode was approximately $100. The new electrodes installed on Telica produced little to no noise. The new electrodes installed on Masaya introduced noise with a frequenc ies above 0.03858 Hz (Figure 9), which is believed to have been due to weaknesses in the connection between the lead wire and copper wire. However, the noise is easily removed with a simple low-pass filter (Figure 13). 20

PAGE 29

Figure 8. Left: Schematic design of Pb-PbCl 2 electrodes from Petiau (2000). Right: Photograph of electrodes construc ted using design by Petiau (2000). Figure 9. Graph of self-potential and rain fall data collected between May, 27, 2006 and March 19, 2007, showing noise added when ne w electrodes were installed on Masaya. 21

PAGE 30

Results SP and CO 2 Flux Profiles Profiles collected on Masaya show ed good correlation between SP and CO 2 flux (Figures 10 and 11). In general peaks in SP corresponded to broad areas of high CO 2 flux, centered over fracture zones. The la rgest amplitude SP anomaly of 82 mV was seen at the Fumarole Field Site, Line 0. However, the largest CO 2 fluxes of 2100 g/m 2 d and 2000 g/m 2 d were recorded at the Fumarole Fi eld Site, Line 2 and the Hilltop Site, respectively. SP profiles tended to be smoother overall than CO 2 flux profiles at all sites. SP CO2 Flux Figure 10. Top: SP and CO 2 Flux profiles from Masaya; Comalito Site, Line 1. Bottom: SP and CO 2 Flux profiles from Masaya; Hilltop S ite. Red pluses are SP measurements and blue triangles are CO 2 flux measurements. 22

PAGE 31

SP CO2 Flux Figure 11. Top: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 0. Center: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 1. Bottom: SP and CO 2 Flux profiles from Masaya; Fumarole Field, Line 2. Red pluses are SP measurements and blue triangles are CO 2 flux measurements. The profile collected on Teli ca shows a much different pattern. There is virtually no SP anomaly and extremely low CO 2 flux for the majority of the profile (Figure 12). The readings began to increase near the end of the profile (nearest to the active crater), with SP values reaching 40 mV and CO 2 flux readings reaching 155 g/m 2 d. The drop in SP values from 34 m to 44 m is due to inte rference from a concrete pad. Unlike the profile collected at Telica, the SP profile at Cerro Negro shows a clear increase toward 23

PAGE 32

the fumarole area at ~20 m, reaching a ma ximum of 115 mV at 36 m (Figure 12). The CO 2 flux also has a maximum of 4300 g/m 2 d at the same point, before returning to the previous level. SP CO2 Flux Concrete Pa d Fumarole Figure 12. Top: SP and CO 2 Flux profiles from Telica. Bottom: SP and CO 2 Flux profiles from Cerro Negro. Red pluses are SP measurements and bl ue triangles are CO 2 flux measurements. Continuous SP Monitoring: Masaya Volcano Representative plots of the data recorded at the continuous monitoring station at Masaya from May 27, 2006 through March 19, 2 007 are shown in Figures 13 through 18. Plots of the entire data set can be found in Appendix 1. The data collected at Masaya show a wide range in both temperature and SP (Figure 13). A persistent positive SP anomaly, which fluctuated between 60 and 240 mV, was seen throughout the monitoring period, even during times of rain fall. Two temperature transi ents, which were recorded 24

PAGE 33

in June and October 2006, are easily seen in the spectrogram for the thermocouple at 90 cm; however they are not as obvious in the sp ectrogram for SP (Figure 14). This may be due in part to the increased noise seen in SP due to the use of the new electrodes. Figure 13. Temperature and rain fall data (top) and SP and ra infall data (bottom) for the period from May 27, 2006 through March 19, 2007 from Masaya. The SP data have been filtered using a low-pass filter to remove frequencies above 0.03858 Hz. Rainfall had the largest effect on the SP anomaly with larger rain events corresponding to the largest dr ops in SP (Figure 15). Rain events with magnitudes of approximately 10 mm or less did not have an effect on the SP anomaly and generally coincided with times of SP increase. A positive SP anomaly of 60 to 220 mV persisted throughout the rainy season with recovery time s after large rain ev ents lasting from several hours to several days, depending on the magnitude of the rain event (Figure 13). 25

PAGE 34

Rainfall also had a stabilizing effect on SP, as diurnal variations were smaller during the rainy season than dur ing the dry season. Figure 14. Spectrograms of SP (top) and temper ature (bottom) from the thermocouple at 90 cm. SP and temperature data used in the spectrograms are unfiltered. Figure 15. Rainfall magnitude versus SP record ed during times of heaviest rainfall in July and October. 26

PAGE 35

The temperature transients in June and October produced the largest increases in temperature seen during the monitoring peri od of 3 9 C and 0.5 4 C, respectively (Figures 16 and 17). Temperature increases during both transients generally coincided with the onset of rain events with only a few exceptions. However, temperature increases did not occur for every rain event throughout the monitoring period. Some of the largest rain events, seen during July, did not corre spond to increases in temperature as seen during June and October (Figure 13). There were also two times, on June 8 9 and June 10, when the temperatures fluctuated during times without rainfall as recorded by the thermocouples at 65, 90, and 150 cm but not by the thermocouples at 33 cm or 7 m away from the system (Figure 16). Despite the large increases in temperature seen during the June and October transients there were no increases record ed in the SP anomaly. Instead the only fluctuations seen in the SP were decreases wh ich coincided with rain events (Figures 16 and 17). There were also several times throughout the monitoring period when significant (27 80 mV) increases in SP were re corded with no apparent change in any other monitored signal (Figure 13). The incr eases occurred around the same time each month from June through January, with larger increases occurring during the dry season. Periodicity during the entire monitoring pe riod in SP and barometric pressure was dominated by 12 and 24 hour periods. Temp erature was also dominated by a 24 hour period, but contained a much less dominant 12 hour period. Rainfall displayed no dominant periods (Figure 18). Periodicit y during the June and October anomalies changed drastically, showing no 12 or 24 hour periods in SP, barometric pressure, or 27

PAGE 36

temperature (Figures 19 and 20). In addition to the diurnal and semi-diurnal periodicity seen, an overall increasing trend was also re corded in temperature and SP (Figure 13). The trend in temperature ranged from 0.004 C/day for the thermocouple at 150 cm to 0.009 C/day for the thermocouple at 65 cm, while the trend in SP was 0.44 mV/day. After the rainy season ended in Janua ry, 2007 the temperature and SP signals began to display a prominent diurnal component with much larger daily variations than had been seen before. An inverse corre lation was seen between SP and temperature during some of the time series (Figures 21, 22, and 23). Peaks in SP were seen to occur approximately two hours before temperat ure lows. Likewise, SP lows occurred approximately two hours before temperature highs. However, when variations in temperature begin to show a longer wavelength the SP signal remains regular (Figure 23). An inverse relationship was also seen betw een diurnal variations in temperature and barometric pressure (Figures 21, 22, and 23). Temperature maxima generally coincided with lows in barometric pressure while temperature minima were seen to coincide with highs in barometric pressure. The temper ature maxima and minima sometimes lagged barometric pressure by a half hour to an hour or did not correlate at all, such as was seen between February 17 through 19 (Figure 23). Small changes in barometric pressure of ~0.3 to 0.6 KPa were seen to have a broad correlation with large ch anges in SP of ~100 mV on a diurnal and semi-diurnal basis, duri ng some parts of the tim es series (Figures 21, 22, and 23). The SP anomalies appear to be slightly out of phase, however, with barometric pressure. Specifically, SP maxi ma precede barometric pressure maxima by approximately two hours. Similarly, SP mini ma precede barometric pressure minima by 28

PAGE 37

about two hours (Figures 21, 22, and 23). A se mi-diurnal component seen in barometric pressure also coincided with small scale fl uctuations in SP (Figures 21, 22, and 23). In the case of this semi-diurnal component, the signals coincide d exactly in phase. Figure 16. Temperature and rainfa ll data (top) and SP and rain fall data (bottom) recorded during the temperature tr ansient in June 2006 at Masaya. Increases in temperature of 3 9 C coincided with the onset of rainfall for all rain events except for the rain event on the morning of June 8. There were no large in creases in SP associated with temperature increases observed during this time period, on ly decreases which coincided with rainfall. 29

PAGE 38

Figure 17. Temperature and rainfa ll data (top) and SP and rain fall data (bottom) recorded during the temperature transien t in October 2006 at Masaya. Increases in temperature of 0.5 4 C coincided with the onset of rainfall for all rain events except for rain events between October 15 17 and on the morni ng of October 25. There were no large increases in SP associated with temperatur e increases observed dur ing this time period, only decreases which coincided with rainfall. 30

PAGE 39

24 hr 12 hr 24 hr 12 hr 24 hr 12 hr Figure 18. Spectral estimates of SP (top left), barometr ic pressure (top right), temperature at the thermocouple at 90 cm ( bottom left), and rainfa ll (bottom right). Showing dominant 12 and 24 hour periods in SP and barometric pressure, a dominant 24 hour period with a much smaller 12 hr compone nt in temperature, and no periods in rainfall. 31

PAGE 40

24 hr 12 hr 24 hr 12 hr 24 hr 12 hr Figure 19. Spectral estimates of SP (top left), barometric pressure (top right), and temperature at the thermocouple at 90 cm (bottom) during the June 2006 temperature transient. Showing no dominant 12 or 24 hour periods in SP, barometric pressure, or temperature. 32

PAGE 41

24 hr 12 hr 24 hr 12 hr 24 hr 12 hr Figure 20. Spectral estimates of SP (top left), barometric pressure (top right), and temperature at the thermocouple at 90 cm (bottom) during the October 2006 temperature transient. Showing no dominant 12 or 24 hour periods in SP, barometric pressure, or temperature. 33

PAGE 42

Figure 21. Temperature at the thermocoupl e at 90cm, barometric pressure, and SP recorded from February 1, 2007 through February 6, 2007 at Masaya. Inverse relationships seen between temperature and SP and between temperature and barometric pressure. SP maxima/minima occur appr oximately two hours before temperature minima/maxima. Temperature maxima/m inima generally coincide with BP minima/maxima, but sometimes lag by one half to one hour. SP and barometric pressure show a broad correlation, w ith SP maxima/minima occur approximately two hours before barometric pressure minima/maxima. The se mi-diurnal component seen in barometric pressure coincides exactly in phase with small scale fluctuations in SP. 34

PAGE 43

Figure 22. Temperature at the thermocoupl e at 90cm, barometric pressure, and SP recorded from February 8, 2007 through February 13, 2007 at Masaya. Inverse relationships seen between temperature and SP and between temperature and barometric pressure. SP maxima/minima occur appr oximately two hours before temperature minima/maxima. Temperature maxima/m inima generally coincide with BP minima/maxima, but sometimes lag by one half to one hour. SP and barometric pressure show a broad correlation, w ith SP maxima/minima occur approximately two hours before barometric pressure minima/maxima. The se mi-diurnal component seen in barometric pressure coincides exactly in phase w ith small scale fluctuations in SP. 35

PAGE 44

Figure 23. Temperature at the thermocoupl e at 90cm, barometric pressure, and SP recorded from February 15, 2007 through February 20, 2007 at Masaya. During breakdowns in the diurnal variations in te mperature the diurnal variations in SP and barometric pressure remain very regular. 36

PAGE 45

Continuous SP Monitoring: Telica Volcano Representative plots of the data recorded at the continuous monitoring station at Telica from June 1, 2006 through March 21, 2007 are shown in Figures 24 and 25. Plots of the entire data set can be found in Appendix 2. Continuous monitoring data collected at Telica displayed much less variation in both SP and temperature than was seen at Masaya (Figure 24). Short term increases in temperature were not seen at Telica, instead most short term temperature changes we re decreases in temperature. Figure 24. Temperature and rain fall data (top) and SP and ra infall data (bottom) for the time period from June 1, 2006 through March 21, 2007 at Telica. Rainfall data are from a rain gauge located ~26 km away from th e volcano. Large varia tions in temperature recorded in June through September at T3 are due to poor wire connections. 37

PAGE 46

Variations of SP and temperature due to rainfall can only be suggested due to the large distance between the rain gauge and the continuous monitori ng station (~26 km). Rainfall did appear to have an effect on SP and temperature at Teli ca, especially during large rain events. However, only rain ev ents on September 1, September 20, and October 5 corresponded to decreases in SP. There was also much less short term variation in both temperature and SP during the dry season. On November 22 there were two anomalous spikes followed by a large drop in SP that corresponded to decreases in temperature at T2 and T3. However, temperatures at T1, the sensor located next to the hot electrode, showed no change in temperature. A clear diurnal and sometimes a semi-diurnal component were seen in the SP data (Figure 25). However, spectral estimates of the temperature data show dominant 24 hour and 11 hour components (Figure 25). Long term trends in temperature were positive, ranging from 0.002 C/day for T1 to 0.025 C/day for T3 (Figure 24). The trend in SP seen at Telica of 0.15 mV/day wa s smaller than the trend seen at Masaya (Figures 13 and 24). 24 hr 12 hr 24 hr 11 hr Figure 25. Spectral estimate of SP and temperature at T1 at Telica, showing dominant 24 and 12 hour periods in SP and 24 and 11 hour periods in temperature. 38

PAGE 47

Discussion The use of the SP method in volcanic area s has recently gained more attention due to successes seen at Unzen (Hashimoto and Tanaka, 1995) and Piton de la Fournaise (Zlotnicki et al., 2001). However, befo re the method can be employed with any confidence as a monitoring and forecasting tool it is important to understand the variability of SP anomalies with time. Thus far, only a few studies have been aimed at understanding these changes (Hashimoto and Tanaka, 1995; Zlot nicki et al., 2001; Friedel et al., 2004). The purpose of this study was first to correlate the SP anomalies seen at Masaya, Telica, and Cerro Negro with CO 2 flux. Secondly, continuous monitoring stations on Masaya and Telica were deployed in order to collect temperature, SP, rainfall, and barometric pressure data w ith the goal of studying patterns in the times series. Given a solid unde rstanding of SP anomalies on volcanoes we can one day employ this approach as a monitoring tec hnique on active volcanoes around the world. SP and CO 2 Flux Profiles SP and CO 2 flux profiles collected on Masaya show good correlation between SP and CO 2 flux highs (Figures 10 and 11). Clear peaks in SP are seen on all profiles completed on Masaya and most likely coincide with fractures where fluid flow is less restricted. The sharp increase in SP seen in the profile completed on Cerro Negro clearly 39

PAGE 48

defines the edge of the fumarole area at 20 m (Figure 12). CO 2 flux measurements at Cerro Negro also show an increase over the fumarole. The profiles completed at Telica showed no SP or CO 2 flux anomalies for the majority of the profile, despite the presence of fumaroles (Figure 12). This suggests that there is very little hydrothermal activity at the profile location. Overall, these observat ion reinforce the idea that SP can indicate areas of diffuse gas flux on the flanks of volcanoes (Lewicki et al., 2003), although the SP anomalies themselves are not the direct result of gas flow. Continuous SP Monitoring: Masaya Volcano Throughout the monitoring period at Masa ya a persistent positive SP anomaly was seen in the time series. This anomaly is most likely due to the electrokinetic potential created by the upwelling of hydrotherm al fluids from depth and from the vadose zone into the fracture zone at Comalito, as is depicted in Figure 26. The amplitude of the anomaly fluctuated between 60 and 240 mV, with the largest fluctuations occurring during the rainy season (Figure 13). The largest SP drops were seen after the largest rain events in July and October (Figures 13 a nd 15). A minimum thresh old of rainfall was needed to produce a response and so the smalle st rain events (less than ~10 mm total rainfall) generally had no observable effect on the SP signal. The recovery time after rain events was generally several hours to severa l days, depending on the size of the rain event. The persistence of the SP anomaly throughout the ra iny season suggests that the rainfall only worked to suppress the anomaly for a short time during and after significant rain events. One explanation for the decrease in SP seen during heavy rainfall is that the 40

PAGE 49

increased amount of recharge during these times likely produced a large increase in pore pressure which in turn caused an increase in the fluid flow through the fracture zone (Figure 27). The increased flow may have c ontributed to an increas e in temperature seen at the ground surface. This temperature contribution, when added to a temperature increase due to volcanic activity, may explain the large temperature increases seen at the onset of rainfall. However, the effects of the downward percolation of the rainfall far exceeded the effects of the increased fluid fl ow in the fracture zone and so worked to suppress the SP anomaly for a short time during and after heavy rain events. The time scale of the suppression was generally several hours to seve ral days and was most likely controlled by the amount of time needed for the additional water to evaporate as water vapor through the fracture system or drain to the water table. The temperature transients seen in Ju ne and October coincided with times of increased activity in Santiago crater, sugges ting a link between the active magma system and the Comalito site. However, significan t increases in SP were not observed during the temperature transients, despite a maximum increase in temperature of 9 C in June and 4 C in October. Instead, large decreases in SP, which coincided with large rain events, were observed during both transients. The model shown in Figure 27 is one explanation for the lack of any significant increase in SP during both temperature transients. However, it also possible that the heat tran sfer mechanism is one which has no effect on SP, such as an influx of heated volcanic gases or fluids with the same ionic charge as the surrounding fluids. 41

PAGE 50

Figure 26. Schematic of the background condi tions at Masaya. Above the groundwater table hydrothermal upwelling from depth or from the vadose z one flows into the fracture zone, which has a higher permeability than the surrounding rock. This upwelling of hydrothermal fluids is recorded at the surf ace as a positive SP anomaly. The temperature of the groundwater at the water table is unknown, but most likely is close to boiling, given vapor flux and fumarole temperatures. The depth to the groundwater table is from MacNeil et al. (in press). The diurnal and semi-diurnal variations s een in temperature, barometric pressure, and SP were much more noticeable during the dry season, but were also present during the rainy season. Closer inspection of the data reveals broad correlations between barometric pressure, SP, and temperature. In the system at Masaya large increases (decreases) in SP (100 mV) occurred approximately two hours before small increases (decreases) in barometric pressure (0.3 0.6 KPa). One explanation for the broad correlation between SP and barometric pressure is that as barometric pressure begins to increase it forces an increase in pore pre ssure, which then produces an increase in the fluid flow into the open fracture zone. This increased flow results in an increase in the 42

PAGE 51

SP anomaly (Figure 28). During times of decreas es in barometric pressure a decrease in pore pressure produces a decrease in the fluid flow into the open fracture zone, resulting in a decrease in the SP anomaly (Figure 29). Changes in pore pre ssure in response to changes in barometric pressure have also b een seen in studies of water wells (Hare and Morse, 1997; Seo, 1997; Rasmussen and Craw ford, 1997; Spane, 2002). It has been shown that in wells and porous media increases in barometric pressure cause increases in pore pressure and decreases in barometric pre ssure cause decreases in pore pressure (Seo, 1997). However, as changes in barometric pressure have opposite effects on SP and temperature, and diurnal varia tions in SP are out of phase with diurnal variations in barometric pressure and temperature it is likely that the generation of the SP anomaly is more complex than the simple model presente d in Figures 28 and 29. One explanation is that the SP anomaly is being driven by some unknown external, diurnal forcing. Another explanation is that the SP anomaly generati on is more sensitive to fluid flow throughout the vadose zone and not just fluid flow from depth through the fracture zone, as appears to be the case for temperature. For example, an introduction of a h eated gas phase into the hydrothermal upwelling from depth would ca use an increase in temperature without affecting the SP anomaly, as was seen during the temperature transients in June and October. In addition, changes in barometric pressure appear to have a different effect on flow through the vadose zone than on flow fr om depth. This is best illustrated during times of increased barometric pressure which show broad correlations with increases in SP but conversely, show broad correlati ons with decreases in temperature. 43

PAGE 52

Figure 27. Schematic of the system at Masaya during times of heavy rainfall. As the increased recharge during times of heavy rain fall enters the system a large increase in pore pressure occurs which increases the fl ow of fluids up the fracture zone. This increased flow may contribute to the increase in temperature seen at the ground surface. A decrease in the SP anomaly is seen because the downward percolation of the heavy rainfall far exceeds the additional upward fluid flow into the fracture zone. The temperature of the groundwater at the water table is unknown, but most likely is close to boiling, given vapor flux and fumarole temperat ures. The depth to the groundwater table is from MacNeil et al. (in press). A significant increasing trend was also seen in temperature and SP data at Masaya throughout the monitoring period. This trend is likely a component of a larger seasonal variation as was seen at Pit on de La Fournaise (Zlotnicki et al., 2001). The increases in SP seen around the same time every month do not correspond to any increases in temperature, suggesting that they are not rela ted to changes in volcanic activity. It is possible that they are associ ated with tidal cycles; howev er more study is needed to determine the exact cause. 44

PAGE 53

Figure 28. Schematic of the system at Masa ya during times of increased barometric pressure. As barometric pressure increases it forces an increase in pore pressure. This increase in pore pressure produces an increa se in the hydrothermal fluid flow into the open fracture zone. The increased flow then contributes to the increase in the SP anomaly. The temperature of the groundwater at the water table is unknown, but most likely is close to boiling, given vapor flux and fumarole temp eratures. The depth to the groundwater table is from MacN eil et al. (in press). 45

PAGE 54

Figure 29. Schematic of the system at Masa ya during times of decreased barometric pressure. As barometric pressure decreases it forces a decrease in pore pressure. This decrease in pore pressure produces a decrease in the hydrothermal fluid flow into the open fracture zone. The decreased flow then contributes to the decrease in the SP anomaly. The temperature of the groundwater at the water table is unknown, but most likely is close to boiling, given vapor flux and fumarole temp eratures. The depth to the groundwater table is from MacN eil et al. (in press). 46

PAGE 55

Continuous SP monitoring: Telica Volcano Unlike data collected at Masaya, data from Telica showed mostly short term decreases in temperature and SP. Many of the short term decreases coincided with rain events, so it is likely that th ese decreases were caused by rainfall. However, because of the large distance from the rain gauge to the continuous monitoring station it is impossible to say for sure which ra in events impacted Telica. Most of the changes in temperature a nd SP were confined to diurnal and semidiurnal variations, which were seen throughout the monitoring period. The only significant changes in SP were the anomalous spikes seen in November. Decreases in temperature in T2 and T3 were recorded dur ing the same time as the anomalous spikes and were likely due to rainfall from a passing storm, which most likely produced lightening. The proximity of the electrodes to a large antenna (Figure 7) raises the susceptibility of the site to lightening st rikes which, coupled with the decrease in temperature in T2 and T3, suggests that the an omalous spikes were due to the passing of a storm. The overall increasing trend seen in the temperature and SP data is most likely part of a larger seasonal variation as at Masaya. The l ack of significant CO 2 flux or increases in SP at Telica despit e reported activity suggests th at the site may experience less hydrothermal activity than at Masaya, which highlights the need for good site selection for SP monitoring. 47

PAGE 56

Conclusion Understanding the phenomena that influence SP anomalies is important to the development of the SP method for use as a volcano monitoring tool. The study presented in this thesis was carried out to investigat e the variations in te mperature, barometric pressure, rainfall, and SP over time. The following conclusions and recommendations can be made on the basis of this study: 1) Spatial surveys identified broad correlations between SP and CO 2 flux in all three areas. These results s upport the notion that short-wave length SP anomalies occur in areas of diffuse degassing on volcanoes as was shown by Lewicki et al. (2003). 2) Large positive SP anomalies of 60 to 240 mV, that occur in these areas of diffuse degassing, have been shown to persist year-round despite seasonal variations in rainfall. 3) Rain events above ~10 mm total ra infall can have a sign ificant short-term effect on SP. These large rain events can produce maximum decreases in SP of 76 mV which can persist for several days to several weeks. SP surveys must either be conducted in the same season, or corrections to the data must be attempted. 48

PAGE 57

4) Broad correlations were seen between small variations in barometric pressure (0.3 0.6 KPa) and very large changes in SP (1 00 mV). The link between variations in barometric pressure and SP shoul d be investigated further. 5) Changes in SP do appear to be connected to mass flow based on spatial surveys of SP and CO 2 flux collected during this and othe r studies (Lewicki et al., 2003) However the relationships between SP, temperatur e, and barometric pressure seen in this study suggest a complex relationship. It is important that correlat ions of SP with mass flow be applied with caution. 6) Rainfall or soil moisture and barome tric pressure should also be monitored during any SP investigation. At Masaya and Telica, a distinctive SP signal was sometimes seen in the data after rain events. The signal consisted of a rapid decrease in SP followed by an increase which approached the background SP signal asymptotically. This distinctive signal may be a characteris tic of SP response to rainfall; however more study is needed to be sure. 7) Site choice is key to the success of any continuous SP monitoring station, as was seen from the data collected at Telic a. Sites should be chosen with minimal elevation change, a large SP anomaly, and ev idence of mass flow (e.g. high volcanic gas flux). Studies of variations in SP anomalie s would also greatly benefit from the use of spatial and temporal continuous monitoring; including the installati on of a network of electrodes at each site, seve ral sites along the flanks of a single volcano, and several studies conducted on multiple volcanoes. 49

PAGE 58

50 References Aizawa, K., 2004, A large self-potential anom aly and its changes on the quiet Mt. Fuji, Japan: Geophysical Research Letters, v. 31, p. 1-4, doi: 10.1029/2004GL019462. Antraygues, P., and Aubert, M., 1993, Self potential generated by two-phase flow in a porous medium; experimental study and vol canological applications: Journal of Geophysical Research, v. 98, p. 22-22,281. Aubert, M., Dana, I.N., and Gourgaud, A., 2000, Internal structure of the Merapi Summit from self-potential measurements; Mera pi Volcano: Journal of Volcanology and Geothermal Research, v. 100, p. 337-343. Bolve, A., Revil, A., Janod, F., Mattiuzzo, J.L., Jardani, A., 2007, A new formulation to compute self-potential signals associated with ground water flow: Hydrology and Earth Systems Sciences Discussions, v 4, p. 1429-1463. Corwin, R.F., and Hoover, D.R., 1979, The self-potential method in geothermal exploration: Geophysics, v. 44, p. 226-245. Di Maio, R., and Patella, D., 1994, Self-poten tial anomaly generation in volcanic areas; the Mt. Etna case-history; The 1991-1993 Etna eruption: Acta Vulcanologica, v. 4, p. 119-124. Dzurisin, D., Anderson, L.A., Eaton, G.P., Koyanagi, R.Y., Lipman, P.W., Lockwood, J.P., Okamura, R.T., Puniwai, G.S., Sako, M.K., and Yamashita, K.M., 1980, Geophysical observations of Kilauea Vol cano, Hawaii; 2, Constraints on the magma supply during November 1975-September 1977; Gordon A. Macdonald memorial volume: Journal of Volcanology and Ge othermal Research, v. 7, p. 241-269.

PAGE 59

51 Ernstson, K., and Scherer, H.U., 1986, Selfpotential variations with time and their relation to hydrogeologic and meteorol ogical parameters: Geophysics, v. 51, p. 19671977, doi: 10.1190/1.1442052. Finizola, A., Lenat, J., Macedo, O., Ramos, D., Thouret, J., and Sortino, F., 2004, Fluid circulation and structural discontinuities inside Misti Volca no (Peru) inferred from selfpotential measurements: Journal of Vol canology and Geothermal Research, v. 135, p. 343-360, doi: 10.1016/j.jvolgeores.2004.03.009. Finizola, A., Sortino, F., Lenat, J., Aubert M., Ripepe, M., and Valenza, M., 2003, The summit hydrothermal system of Stromboli. New insights from self-potential, temperature, CO 2 and fumarolic fluid measurements with structural and monitoring implications: Bulletin of Volcanology, v. 65, p. 486-504. Finizola, A., Sortino, F., Lenat, J., and Vale nza, M., 2002, Fluid circulation at Stromboli Volcano (Aeolian Islands, Italy) from selfpotential and CO (sub 2) surveys: Journal of Volcanology and Geothermal Research, v. 116, p. 1-18. Friedel, S., Byrdina, S., Jacobs, F., and Zimmer, M., 2004, Self-potential and ground temperature at Merapi Volcano prior to its crisis in the ra iny season of 2000-2001: Journal of Volcanology and Geothe rmal Research, v. 134, p. 149-168, doi: 10.1016/j.jvolgeores.2004.01.006. Guichet, X., Jouniaux, L., and Pozzi, J., 2003, Streaming potential of a sand column in partial saturation conditions: Journal of Geophysical Research, v. 108, p. 12, doi: 10.1029/2001JB001517. Hare, P.W., and Morse, R.E., 1997, Water-level fluctuations due to barometric pressure changes in an isolated portion of an unc onfined aquifer: Ground Water, v. 35, p. 667-671. Hase, H., Ishido, T., Takakura, S., Hashim oto, T., Sato, K., and Tanaka, Y., 2003, zeta potential measurement of volcanic rocks from Aso Caldera: Geophysical Research Letters, v. 30, p. 4, doi: 10.1029/2003GL018694. Hashimoto, T., and Tanaka, Y., 1995, A large self-potential anomaly on Unzen Volcano, Shimabara Peninsula, Kyushu Island, Japan: Geophysical Research Letters, v. 22, p. 191194.

PAGE 60

52 Ishido, T., 1989, Self-potential generation by subsurface water flow through electrokinetic coupling; Detection of subsurface flow phenomena by selfpotential/geoelectrical and th ermometric methods: Lecture Notes in Earth Sciences, v. 27, p. 121-131. Ishido, T., and Pritchett, J.W ., 1999, Numerical simulation of electrokinetic potentials associated with subsurface fluid flow: J ournal of Geophysical Research, v. 104, p. 1515,259. Jackson, D.B., and Kauahikaua, J.P., 1987, Regi onal self-potential a nomalies at Kilauea Volcano; Volcanism in Hawaii: U. S. Ge ological Survey, Reston, VA, United States (USA), Report P 1350, 947-959 p. Johnston, M.J.S., Byerlee, J.D., and Lockner, D.A., 2001, Rapid fluid disruption; a source for self-potential anomalies on volcanoes: Journal of Geophysical Research, v. 106, p. 4327-4335. Johnston, M.J.S., Lockner, D.A., and Byerlee, J.D., 2002, Rapid fluid disruption; a source for self-potential anomalies on volcanoes; reply [modified]: Journal of Geophysical Research, v. 107, p. 7, doi: 10.1029/2002JB001794. Kanda, W., and Mori, S., 2002, Self-potential anomaly of Satsuma-Iwojima Volcano; Satsuma-Iwojima; continuous degassing of a rhyolitic volcano: Earth, Planets and Space, v. 54, p. 231-237. Lewicki, J.L., Connor, C., St-Amand, K., Sti x, J., and Spinner, W ., 2003, Self-potential, soil CO (sub 2) flux, and temperature a nd Masaya Volcano, Nicaragua: Geophysical Research Letters, v. 30, p. 4, doi: 10.1029/2003GL017731. Lewicki, J.L., Hilley, G.E., and Connor, C., 2004, The scaling relationship between selfpotential and fluid flow on Masaya Volca no, Nicaragua; Proceedings of the Eleventh international symposium on Water-rock inte raction: Proceedings International Symposium on Water-Rock In teraction, v. 11, p. 153-156.

PAGE 61

53 Macneil, R.E., Sanford, W., Connor, C.B., Sandber, S.K., Diez, M., 2006, Investigation of the groundwater system at Masaya Caldera, Nicaragua using transient electromagnetics and numerical simulation: Journal of Volcanology and Geothermal Research, in press. Massenet, F., and Ngoc, P.V., 1985, Mapping an d surveillance of ac tive fissure zones on a volcano by the self-potential method, Etna, Sicily: Journal of Volcanology and Geothermal Research, v. 24, p. 315-338. Michel, S., and Zlotnicki, J., 1998, Self-poten tial and magnetic surveying of La Fournaise Volcano (Reunion Island); correlations with faulting, flui d circulation, and eruption: Journal of Geophysical Research, v. 103, p. 17-17,857. Morgan, F.D., 1989, Fundamentals of streami ng potentials in geophys ics; laboratory methods; Detection of subsurface flow pheno mena by self-potential/geoelectrical and thermometric methods: Lecture Notes in Earth Sciences, v. 27, p. 133-144. Nourbehecht, B., 1963, Irreversible thermodynamic effects in inhomogeneous media and their application in certain geoelectric problems [Ph.D. thesis]: United States (USA), Massachusetts Institute of Technology, Cambridge, MA, United States (USA). Perrier, F., and Morat, P., 2000, Characterizati on of electrical daily variations induced by capillary flow in the non-saturated zone : Pure and Applied Geophysics, v. 157, p. 785810. Petiau, G., 2000, Second genera tion of lead-lead chloride electrodes for geophysical applications: Pure and Applied Geophysics, v. 157, p. 357-382. Rasmussen, T.C., and Crawford, L.A., 1997, Identifying and removing barometric pressure effects in confined and unconfin ed aquifers: Ground Water, v. 35, p. 502-511. Revil, A., 2002, Rapid fluid disruption; a source for self-potential anomalies on volcanoes; discussion [modifi ed]: Journal of Geophysical Research, v. 107, p. 4, doi: 10.1029/2001JB000788.

PAGE 62

54 Revil, A., and Cerepi, A., 2004, Streaming potentials in two-phase flow conditions: Geophysical Research Letters, v. 31, p. 4, doi: 10.1029/2004GL020140. Revil, A., Hermitte, D., Spangenberg, E., a nd Cocheme, J.J., 2002, Electrical properties of zeolitized volcaniclastic materials: Jour nal of Geophysical Research, v. 107, p. 17, doi: 10.1029/2001JB000599. Revil, A., Pezard, P.A., and Glover, P.W.J., 1999, Streaming potential in porous media; 1, Theory of the zeta potential: Journa l of Geophysical Research, v. 104, p. 20-20,031. Revil, A., Saracco, G., and Labazuy, P., 2003, The volcano-electric effect: Journal of Geophysical Research, v. 108, p. 20, doi: 10.1029/2002JB001835. Revil, A., Schwaeger, H., Cathles, L.M .,III, and Manhardt, P.D., 1999, Streaming potential in porous media; 2, Th eory and application to geot hermal systems: Journal of Geophysical Research, v. 104, p. 20-20,048. Revil, R., and Pezard, P.A., 1998, Streaming elec trical potential anomaly along faults in geothermal areas: Geophysical Re search Letters, v. 25, p. 3197-3200. Roche, O., van Wyk de Vries,B., and Dr uitt, T.H., 2001, Sub-surface structures and collapse mechanisms of summit pit craters: Journal of Volcanology and Geothermal Research, v. 105, p. 1-18. Sasai, Y., Uyeshima, M., Zlotnicki, J., Ut ada, H., Kagiyama, T., Hashimoto, T., and Takahashi, Y., 2002, Magnetic and electric fi eld observations duri ng the 2000 activity of Miyake-Jima Volcano, central Japan: Earth and Planetary Science Letters, v. 203, p. 769777. Sato, M., and Mooney, H.M., 1960, The electr ochemical mechanism of sulfide selfpotentials: Geophysics, v. 25, p. 226-249. Schuch, M., 1989, Streaming potential in na ture; Detection of subsurface flow phenomena by self-potential/geoelectrical and thermometric methods: Lecture Notes in Earth Sciences, v. 27, p. 99-107.

PAGE 63

55 Seo, H.H., 2001, Modeling the influence of changes in barometric pressure on groundwater levels in wells: Environmental Geology (Berlin), v. 41, p. 155-166. Sill, W.R., 1983, Self-potential modeling from primary flows: Geophysics, v. 48, p. 7686, doi: 10.1190/1.1441409. Spane, F.A., 2002, Considering barometric pressure in groundwater flow investigations: Water Resources Research, v. 38, p. 18, doi: 10.1029/2001WR000701. Tort, A., and Finizola, A., 2005, The buried caldera of Mist i Volcano, Peru, revealed by combining a self-potential su rvey with elliptic Fourier f unction analysis of topography: Journal of Volcanology and Geothe rmal Research, v. 141, p. 283-297, doi: 10.1016/j.jvolgeores.2004.11.005. Zlotnicki, J., Sasai, Y., Yvetot, P., Nishid a, Y., Uyeshima, M., Fauquet, F., Utada, H., Takahashi, Y., and Donnadieu, G., 2003, Resistiv ity and self-potential changes associated with volcanic activity; the July 8, 2000 Mi yake-Jima eruption (Japan): Earth and Planetary Science Letters, v. 205, p. 139-154. Zlotnicki, J., Le Mouel, J., Sasai, Y., Yv etot, P., and Ardisson, M.H., 2001, Self-potential changes associated with volcanic activity; shor t-term signals associated with March 9, 1998 eruption on La Fournaise Volcano (Reunion Island); Proceedings of the 2nd international workshop on Magnetic, electric and electromagnetic methods in seismology and volcanology: Annali Di Ge ofisica (1993), v. 44, p. 335-354. Zlotnicki, J., and Nishida, Y., 2003, Review on morphological insights of self-potential anomalies on volcanoes: Surveys in Geophysics, v. 24, p. 291-338.

PAGE 64

56 Appendices

PAGE 65

Appendix 1: Addition al Plots, Masaya The following are plots of the entire data set created during the analysis of data collected at the continuous monitoring station at Masaya between May 27, 2006 and March 19, 2007. 24 hr 12 hr Figure 1. Spectral estimate of self-potential at Masaya, showi ng dominant 12 and 24 hour periods. 57

PAGE 66

Appendix 1 (continued) 24 hr 12 hr Figure 2. Spectral estimate of barometric pr essure at Masaya, showing dominant 12 and 24 hour periods. 24 hr 12 hr Figure 3. Spectral estimate of temperature at 90 cm at Masaya, showing a dominant 24 hour period and minor 12 hour period. 58

PAGE 67

Appendix 1 (continued) Figure 4. Spectral estimate of rainfall at Masaya, showing no dominant periods. 24 hr 12 hr Figure 5. Spectral estimate of self-potential during the June anomaly at Masaya, showing no dominant 12 or 24 hour periods. 59

PAGE 68

Appendix 1 (continued) 24 hr 12 hr Figure 6. Spectral estimate of barometric pressure during the June anomaly at Masaya, showing no dominant 12 or 24 hour periods. 24 hr 12 hr Figure 7. Spectral estimate of temperature at 90 cm during the June anomaly at Masaya, showing no dominant 12 or 24 hour periods. 60

PAGE 69

Appendix 1 (continued) 24 hr 12 hr Figure 8. Spectral estimate of self-potential during the October anomaly at Masaya, showing no dominant 12 or 24 hour periods. 24 hr 12 hr Figure 9. Spectral estimate of barometric pressure during the October anomaly at Masaya, showing no dominant 12 or 24 hour periods. 61

PAGE 70

Appendix 1 (continued) 24 hr 12 hr Figure 10. Spectral estimate of temperature at 90 cm during the October anomaly at Masaya, showing no dominant 12 or 24 hour periods. 62

PAGE 71

Appendix 1 (continued) Figure 11. Temperature and rain fall (top) and self-potential and rainfall (bottom) fo r time period from May 27, 2006 through March 21, 2007 for Masaya. SP data have been filtered using to low-pass filter. 63

PAGE 72

Appendix 1 (continued) Figure 12. Temperature and rain fall (top) and self-potential and rainfall (bottom) fo r time period from May 27, 2006 through March 21, 2007 for Masaya, show ing long term trends. SP data ha ve been filtered using a low-pass filter. 64

PAGE 73

Appendix 1 (continued) Figure 13. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from June 4, 2006 through June 30, 2006 for Masaya. 65

PAGE 74

Appendix 1 (continued) Figure 14. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from June 4, 2006 through June 13, 2006 for Masaya. 66

PAGE 75

Appendix 1 (continued) Figure 15. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from June 7, 2006 through June 8, 2006 for Masaya. 67

PAGE 76

Appendix 1 (continued) Figure 16. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from June 9, 2006 through June 10, 2006 for Masaya. 68

PAGE 77

Appendix 1 (continued) Figure 17. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from June 11, 2006 through June 12, 2006 for Masaya. 69

PAGE 78

Appendix 1 (continued) Figure 18. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from October 15, 2006 throug h October 28, 2006 for Masaya. 70

PAGE 79

Appendix 1 (continued) Figure 19. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from October 18, 2006 throug h October 19, 2006 for Masaya. 71

PAGE 80

Appendix 1 (continued) Figure 20. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from October 20, 2006 throug h October 21, 2006 for Masaya. 72

PAGE 81

Appendix 1 (continued) Figure 21. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from October 22, 2006 throug h October 23, 2006 for Masaya. 73

PAGE 82

Appendix 1 (continued) Figure 22. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from October 24, 2006 throug h October 25, 2006 for Masaya. 74

PAGE 83

Appendix 1 (continued) Figure 23. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from July 6, 2006 through July 25, 2006 for Masaya. 75

PAGE 84

Appendix 1 (continued) Figure 24. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from September 6, 2006 through September 27, 2006 for Masaya. 76

PAGE 85

Appendix 1 (continued) Figure 25. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from January 28, 2007 through February 27, 2007 for Masaya. 77

PAGE 86

Appendix 1 (continued) Figure 26. Temperature and rainfa ll (top) and self-potential a nd rainfall (bottom) for time period from February 8, 2007 throug h February 10, 2007 for Masaya. 78

PAGE 87

Appendix 1 (continued) 79 Figure 27. Temperature difference and rainfall (top) and self-potential and rainfall (bottom) for time period from May 27, 2006 through March 21, 2007 for Masaya. SP data have been filtered us ing a low-pass filter.

PAGE 88

Appendix 1 (continued) Figure 28. Temperature differen ce and rainfall (top) and self-potential and rainfall (bottom) for time period from June 5, 2006 through June 11, 2006 for Masaya. 80

PAGE 89

Appendix 1 (continued) Figure 29. Temperature differen ce and rainfall (top) and self-potential and rainfall (bottom) for time period from October 15, 2006 through October 28, 2006 for Masaya. 81

PAGE 90

Appendix 1 (continued) Figure 30. Barometric Pressure (top) and self -potential (bottom) for time period from November 11, 2006 through March 19, 2007 for Masaya. 82

PAGE 91

Appendix 1 (continued) 83 Figure 31. Barometric Pressure (top) and self -potential (bottom) for time period from February 8, 2007 through February 10, 2007 for Masaya.

PAGE 92

Appendix 2: Additional Plots, Telica The following are plots of the entire data set created during th e analysis of data collected at the continuous monitoring stat ion at Telica between June 1, 2006 and March 21, 2007. 24 hr 12 hr Figure 1. Spectral estimate of self-potential at Telica, showi ng dominant 12 and 24 hour periods. 84

PAGE 93

Appendix 2 (continued) 24 hr 11 hr Figure 2. Spectral estimate of temperature at T1 at Telica, showing dominant 11 and 24 hour periods. 85

PAGE 94

Appendix 2 (continued) 86 Figure 3. Temperature and rainfall (top) and self-potential and rainfall (bottom) for Telica Volcano.

PAGE 95

Appendix 2 (continued) 87 Figure 4. Temperature and rainfall (top) and self-potential and rain fall (bottom) for Telica, showing long term trends.

PAGE 96

Appendix 2 (continued) Figure 5. Temperature (top) and self-pot ential (bottom) for time period from 88 October 8, 2006 through October 30, 2006 for Telica.

PAGE 97

Appendix 2 (continued) Figure 6. Temperature (top) and self-pot ential (bottom) for time period from 89 November 10, 2006 through November 30, 2006 for Telica.

PAGE 98

Appendix 2 (continued) Figure 7. Temperature (top) and self-pot ential (bottom) for time period from 90 January 4, 2007 through January 18, 2007 for Telica.

PAGE 99

Appendix 2 (continued) Figure 8. Temperature (top) and self-pot ential (bottom) for time period from 91 March 10, 2007 through March 20, 2007 for Telica.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001920230
003 fts
005 20080107130041.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080107s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002108
035
(OCoLC)187972868
040
FHM
c FHM
043
ncnq---
049
FHMM
090
QE26.2 (ONLINE)
1 100
Lehto, Heather L.
0 245
Self-potential anomalies and CO2 flux on active volcanoes :
b insights from time and spatial series at Masaya, Telica, and Cerro Negro, Nicaragua
h [electronic resource] /
by Heather L. Lehto.
260
[Tampa, Fla.] :
University of South Florida,
2007.
520
ABSTRACT: Considerable effort worldwide has gone into monitoring heat and mass transfer at active volcanoes, as this information may provide clues about changes in volcanic activity and impending eruptions. One method used is the self-potential (SP) method, which has been employed on volcanoes to map hydrothermal systems and structural features and to monitor changes in the hydrothermal system due to volcanic activity. Continuous monitoring of SP has been employed on a few volcanoes and has produced encouraging results. This study presents new time series data collected from continuous monitoring stations at Masaya and Telica, and spatial series data from Masaya, Telica, and Cerro Negro, three active volcanoes in Nicaragua. The primary goals of this study were to determine whether correlations between SP anomalies and CO2 flux exist and to investigate temporal variations in temperature, SP, rainfall, and barometric pressure.To achieve these goals, SP and CO2 flux surveys were conducted on Masaya, Telica, and Cerro Negro, and continuous monitoring stations were installed on Masaya and Telica. The continuous monitoring station on Masaya recorded temperature, SP, rainfall, and barometric pressure. The station on Telica recorded temperature and SP. Profiles collected on Masaya and Cerro Negro show broad correlation between SP and CO2 flux. However, profiles on Telica revealed virtually no SP anomaly or CO2 flux for the majority of the profile, at the time of data collection. Data collected from the continuous monitoring station at Masaya showed a persistent positive SP anomaly that fluctuated between 60 and 240 mV. Rainfall was seen to supress the anomaly for time scales of several hours to several days. Correlations between temperature, SP, and barometric pressure were also seen at Masaya.Curiously, no increases in SP were seen during two temperature transients that occurred during volcanic activity in June and October. Continuous monitoring data from Telica showed only decreases in temperature and SP, which coincided with rainfall. The continuous monitoring data collected in this study and others have begun to provide a better understanding of the nature of SP anomalies, which may aid in the development of the SP method as a volcano monitoring tool.
502
Thesis (M.S.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 91 pages.
590
Advisor: Charles Connor, Ph.D.
653
Hydrothermal system.
Fluid flow.
Diffuse degassing.
Mass flow.
Continuous monitoring.
690
Dissertations, Academic
z USF
x Geology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2108