Atypical Variations of Water Conductivity Prior to Tectonic Earthquakes

Within the framework of precursor events related to earthquakes, this paper analyzes the possible effect on the aquatic environment of the surrounding energy that accompanies earthquakes, particularly in the area where oceanic and continental plates collide (Cocos Plate and North American Plate, south of Mexico). As a preamble, the types of precursor events, characteristics, and their possible origin are described. A project was designed under the assumption that in areas with high frequency and intensity seismicity there is an electrical and electromagnetic potential promoter which is detectable and assessable indirectly by measuring water conductivity behavior, which also may have atypical variations of data; the outcome of intensive conductivity monitoring in different settings, natural as well as manmade (wellsprings, artesian well and a cistern), are presented herein. The results of the conductivity monitoring for seven months, highlight two patterns in data behavior: one pattern shows the subtle dependence of data behavior on the geographic location of data monitoring instruments, revealing that could have a slight relationship between areas with increased seismic frequency and intensity and the presence of atypical conductivity variations. Another pattern reveals the possible relationship between atypical variations in conductivity and subsequent earthquake events; a total of 241 seismic events were analyzed and 59 of them are provided as evidence related with patterns mentioned.


Introduction
Frequently, phenomena linked to earthquake are associated with the interaction of the electric and electromagnetic energy emanating from the Earth's interior with surrounding natural environment.The power of energy ionization upon reaching the surface through faults or unstable areas is manifested by low frequency electromagnetic waves [1]- [5]; alterations of the local ionosphere [6]- [14]; luminous phenomena in the local atmosphere [15]- [17]; and the formation of aerosols and clouds [1] [2] [4] [5] [18]- [20].Energy flow and electromagnetic fields also strengthen and promote the attraction of surrounding radiation issuing from each fracture of rock, thus intensifying the presence of ionizing energy, similar to the phenomenon that occurs near power transmission lines [21] [22].
Usually, these types of phenomena are associated with various origins: one is apparently related to the internal dynamics of the Earth's deep layers which cause electric and electromagnetic energy pulses that interact with the external layers by lithosphere fractures [1] [2] [18].Another possible origin is mechanical and relates to the endogenous forces generated in the fault zones, particularly due to the movement of the Earth's crust [2] [23]; these phenomena have also been associated with the natural decay of unstable radioactive isotopes such as uranium, radium, radon, polonium, lead, and bismuth [7]- [9] [11] [12] [24]- [26].A further cause, also mechanical in nature, is linked with endogenous forces that develop in the tectonic faults and their effect at the molecular level, specifically on the minerals that make up the rock, which results in the formation of an electric and electromagnetic potential flowing within a radius of several kilometers [27]- [32].This energy has the ability to interact with surrounding water bodies, affecting the behavior of terrestrial and aquatic life prior to and subsequent to an earthquake [31]- [33].A recently published example is the behavioral changes found in some amphibians due to disturbances among K + , Na + , Cl − , and Ca 2+ , ions present both inside and outside the cell membrane [31] [32]; the presence of an external electric field triggers an ionic oscillation or vibration which exerts a strong internal pressure in terms of electric charge [34].
Based on the above considerations, research relying on two hypotheses was designed.The first one states that such electric and electromagnetic potential is indirectly detectable and calculated with conductivity measuring instruments, particularly in areas where oceanic and continental plates collide and where seismic activity is frequent and intense.The second one is that the natural energy emanating from the surrounding environment (electromagnetic field and electric potential) could interfere with the normal functioning of the instruments being used, causing unusual variations in conductivity measurements, possibly by interrupting ionic conduction.According with this, a series of measurements to monitor conductivity in aquatic environments were performed in two different settings: a natural setting consisting of wellsprings, and a semi controlled setting using a cistern (tank) and an artesian well.

Study Area Description
After Japan, Indonesia, Chile, Papua New Guinea and Turkey, Mexico is one of the countries with the most seismic activity in the world, and data about its territory and its history on the subject dates from 1460 to the present day [35], Figure 1.Based on this frame of reference, monitoring sites were selected within the states of Guerrero, Oaxaca, and Morelos close to areas with a significant seismic history.These sites were located in the C and D seismic regions of the Mexican national territory (Figure 1), associated with the subduction process of the Cocos Plate, which moves beneath the North American Plate at a rate of six centimeters per year, and shifts 300 kilometers almost horizontally to enter the mantle with a 55˚ inclination angle [36].
For the development of this research, painstaking monitoring of the conductivity was carried out in an aquatic environment by a brief series of time, monitoring 90 days using two different settings: a natural setting, consisting of wellsprings and, a semi artificial setting using a cistern and an artesian well.The chosen wellsprings are located at coordinates 17˚33.606'N,99˚24.583'Wand 18˚28.697'N,99˚9.209'W, Figure 2 and Table 1.The first site is located in the town of Tixtla in the state of Guerrero, Mexico, and it belongs to a recreational center called "Teoixtla" in an area having a geology of sedimentary volcanic rock and gypsum from the Neogene period [37].The second site is located in the town of Tlaquiltenango in the state of Morelos, Mexico.It is also a recreational center, called "Los Manantiales", having a geology of Paleogene rhyolites [38].The semi artificial site corresponds to an artesian well of 1 × 1 × 6 meter located at the geographic coordinates 16˚14.180'N,98˚11.406'W(Figure 2) in the community of "Corralero", a municipality of Pinotepa Nacional in the state of Oaxaca, Mexico, site located in a region with the largest number of earthquakes per square kilometer.The geomorphology and   Seismic events used corresponded to 241: 25 events for "Teoixtla", 9 events for "Los Manantiales," 101 events for "Cistern", and 106 events for "Corralero".

Materials and Methods
Intense monitoring in the sites above mentioned ware conducted using by submersion, a HOBO U24-001 Data Logger to determine conductivity (µS/cm), complementarily a HI 9828 multiparameter meter was used, to determine water quality (Figure 3).Each site was continuously monitored at a sampling frequency of 45 seconds on average, representing a total collection of 18,550 data per monitoring period obtained at least during two weeks per each campaign.Readings of basic water quality parameters of ˚C, pH, Dissolved Oxygen (ppm), and salinity were determined every second with the HI 9828 device for a period of 30 minutes, from which the average value was calculated; monitoring periods for each of the settings are included in Table 1.
Information of all >Mw3.0 earthquakes occurred within each monitoring period was gathered, particularly within the territorial limits of the states of Guerrero and Oaxaca, Mexico.Data was obtained from the National Seismological Service (SSN: http://www2.ssn.unam.mx:8080/index.html) of Mexico.Concerning the wellsprings, only information of seismic events located within a radius not greater than 220 kilometers was gathered.Seismic events obtained for the cistern were located at a distance not greater than 320 kilometers.In both cases, the magnitudes of the earthquakes were normally not higher than 4.5 degrees; however, any earthquake with a magnitude of 5 or greater was addressed regardless of its distance from the location where the conductivity Data Logger had been installed.Regarding the seismic data obtained for the artesian well, the distance was not greater than 50 kilometers, taking in a count that this site is located in a place with the highest seismicity of the region.The general criteria employed to vary the distances was the instrument's geographic position in relation to seismic events occurring in the surrounding area as well as the speed and distance of seismic wave propagation in the ground and the area of influence.
Tabulated data of seismic events obtained from SSN were imported to a geographic information system (GIS) for spatial distribution.The position in geographic coordinates of the Data Logger were taken as reference to filter seismic data according to the place of monitoring (Table 2), and the distance of each seismic event was determined in relation to the Data Logger position.The resulting table was edited, removing events located in  the marine environment as well as those located beyond the distances set in the aforementioned criterion.Information was plotted for each of the locations listed in Table 2, using the 18,550 conductivity data obtained with the HOBO conductivity U24-001.Every seismic event shown in the list resulting from the filtering process was positioned in each chart according to the day and time it occurred; this process was replicated for each of the monitored locations.

Results
For the wellsprings, cistern and the artesian well, variations in the frequency of the monitoring sampling did not represent any significant changes at the time of processing and plotting the results.Another aspect that distinguished the chosen sites, besides their geological conditions, was their geographical position (Figure 2), both that of the instruments as that of the seismic events that occurred in the surrounding area.The monitoring sampling was during a 90 day period, from September 22, 2014 to March 1, 2015: 18 days corresponded to the "Teoixtla" site, 6 to the "Los Manantiales" site, 24 to the "Cistern" site, and 42 to the "Corralero" site, although they were short periods of time it was detailed monitoring.The assessment period for the "Los Manantiales" site was the shortest with one week; more monitoring time was considered unnecessary due to the presence of excessive "interference or noise" detected at the time of plotting data (Figure 4).Noisiness may have been due to two possible causes: a) the Data Logger operation was interfered with by a strong drift or movement, the device being located in a place with an intense current flow; or b) the possible presence of an underground river located relatively close to the wellspring being monitored, the undercurrent being located at approximately 1500 meters, its influence could be due to the continuous aquifer recharge and the presence of ions and possibly radioactive gases like radon [5] [39]- [41].The most intensive sampling took place in the location called "Corralero" as this site is characterized by intense seismic activity.Historically, the surrounding area is where the greatest number of earthquakes per square kilometer occurs, Figure 1.During monitoring, seismic events occurred as close as one kilometer away from monitoring instruments.The location of each seismic event in the Data Logger chart showed an interesting series of patterns.The first pattern observed was the dependence of the atypical readings of conductivity on the geographic location of the monitoring instruments, revealing that there is a possible relationship between areas with greater seismic frequency and intensity and the occurrence of higher atypical conductivity drops behaving such as pulses.As the distance between the Data Logger and the area with the greatest seismic frequency and intensity was reduced, data showed more atypical and more defined conductivity changes.Therefore, the spatial variation of conductivity (µS/cm) at the "Los Manantiales" site has a very confusing pattern (Figure 4), it was very regular at the  3), and it had a very erratic character at the "Teoixtla" site (Figures 9-11).The "Corralero" site showed the greatest variation, having a more defined pattern as well as an average of seven conductivity incidents per week, Figures 12-17.
Another pattern reveals the probable relationship between changes or atypical conductivity drops and the subsequent occurrence of earthquakes.According to these two patterns, one of the most notably changes observed occurred first at the "Teoixtla" site (Figures 9-11 and Table 4).
Based on the two patterns observed, the "Corralero" site differed from the other sites used for monitoring conductivity; one similar patterns had already been observed at the "Teoixtla" site described in previous paragraphs.Variations in conductivity measurements were more perceptible at the "Corralero" site as the Data Logger was closer to the area with the greatest number of earthquakes per square kilometer.During the week of January 19 to 25, six major variations occurred, two of them very close to each other (21/01/15); five of them preceded five seismic events.During the week of January 26 to February 1, six variations occurred; five of them were preceded by six seismic events.Details of atypical conductivity drops in these periods are shown in Figure 12, Figure 13 and Table 5.
During the weeks from February 3 to 9 seven of atypical conductivity behaviors occurred, respectively, as shown in the Figure 14 and three variations occurred in a single day (07/02/15).In the first week, Figure 14, three of the variations were preceded by seismic events; from February 10 to 15, four atypical variations occurred, two of which were preceded immediately by seismic events, Figure 15 and Table 6.
For the period of the week from February 16 to 22, nine atypical readings occurred with three major preceding seismic events, Figure 16.During the week from February 23 to March 1st, nine atypical readings occurred, four of which were preceded by four nearby earthquakes, Figure 17 and Table 7.At the "Corralero" site, aver-               age values of water quality obtained with the HI 9828 instrument were: 27.76 Temperature ˚C, 7.13 pH; 1.17 Dissolved Oxygen ppm, and 0.33 salinity.

Discussion
A total of 241 seismic events were analyzed and 59 of them were associated with atypical conductivity readings: 57 were superficial seismic events, and two were intermediate events; these latter with epicenters located 80 kilometers and 66 kilometers deep, respectively, but some quite distant from the Data Logger, 279 kilometers and 154 kilometers away, correspondingly.As expected, out of the 59 seismic events, 36 occurred in the Corralero site within the surroundings of the most seismically active area in the country, near the towns of Ometepec, state of Guerrero, and Pinotepa Nacional, state of Oaxaca, Mexico, likewise the notable conductivity variations occurred in this place (Figures 12-17).Energy manifestations and phenomena that occur before and after an earthquake can be grouped into two major categories: those events associated with the MECHANICAL ENERGY when blocks of rocks break their position of equilibrium, in this set are included events as terrain deformations or tilting [42]; changes in subsurface porosity, changes in water levels, as well as the release and presence of radioactive gases such as Radon 222 in the surface [2] [5] [43]- [47].Other category is related to the IONIZING ENERGY, involve energy processes before the earthquake occurs, fog and cloud formation in areas near fault lines being among those frequently mentioned [1] [6] [18]; too changes in atmospheric electricity, temperature, and relative humidity [8] [24] [49] [50]; the presence of electromagnetic flux [48]; lights seen over ridges and mountaintops [16] [50]; and changes in animal and human behavior [2] [31] [32] [51]- [54].All of these phenomena manifestations provide strong evidence of energy existence associated with earthquakes, but not of their exact origin.However, they should not be considered as merely curious, amusing, or even entertaining phenomena, such as the presence of lights or glares before and after an earthquake.Such phenomena should be of scientific interest because of the effects that predecessor earthquakes and their accompanying events have on the natural environment and on human settlements.
Atypical conductivity drops at the "Corralero" site (Figures 12-17) behaving as pulses, may provide indirect evidence of the presence of surrounding energy as presumed and mentioned in previous paragraphs.The subtle relationship between conductivity variations and earthquakes in aquatic environments near seismic areas and the energy emanating from the Earth's crust are evidence meant to further contribute to the understanding of these events and their preceding accompanying phenomena.As stated in the second assumption supporting this research, natural energy emanating from the surrounding environment would be sufficient to interfere with the normal operation of the Data Logger, interrupting the electric current or ionic conduction, causing unusual variations in conductivity; apparently, this interference appears to be a fact.This anomaly leads to two more assumptions: ion movement was directed towards a field of more intense energy, or the electric flux between the electrodes of the appliance was interfered with by the same field of natural energy (electromagnetic fields), altering the movement of ions.
In order to find a supported answer to these variations, one possibility mentioned in the literature is that changes in electric conductivity may be caused by the mixing (recharge) of different qualities of water because of the increases or decreases of ion concentration [55] or possibly also is due also to terrain compress or tilting [42] caused by very light earthquakes which increase internal water flow rising mixing.Mixing by natural processes of ground-water and surface-water interaction appears to be unsupported since the water of the "Corralero" site well does not seem to have an exchange with or to be renewed by a nearby wellspring or rainwater because sampling was conducted during the dry season of the year.An additional factors might be the temperature or the proximity of a coastal lagoon, inclusively with all due caution, others could be the Coriolis effect caused by Earth rotation movement or the Lunar influence by gravitational attraction, some of these possible answers would be matter of research.Indisputable conductivity variations do not follow a set pattern, as shown in the sampling site "Corralero", from week to week the number of variations is not constant, in a few hours can be presented two o three variations and the next week be more regular with one per day, Figure 12 and Figure 13.In another period it can present a week with two very close variations in a few hours and the next week it can increase the number of variations, appearing two or three of them in very short periods of time, Figure 16 and Figure 17.
A common element to the ions' change of direction or the interference of the instruments' electric field could be the presence of an electromagnetic field in the surrounding environment; in this case, electromagnetic pulses emanating from the Earth's crust or deeper layers, evidence of which is an anomaly in conductivity (Figures 12-17).This electromagnetic energy has often been detected and considered a precursor of seismic events by various authors and it is considered one of the anomalies with the greatest time interval of environment permanency; furthermore, it is characterized by an unusual increase of the electric potential near epicenters [3] [4] [7] [8] [12] [13] [48].Therefore, the systematic monitoring of conductivity, preferably established in areas with frequent and intense seismic activity, may contribute more to understanding of this type of phenomenon.For this reason, it is a sustainable option and can be a preventive tool when considering the apparent relationship existing between water conductivity and subsequent earthquake occurrences.

Conclusions
According to the results described in this paper is concluded: • The conductivity monitoring process of water in the wellsprings, the cistern, and the artesian well provide evidence of the possible existence of local energy that is manifested with drops or atypical conductivity readings in aquatic environments close to areas of frequent seismic events; as well as of the possible relationship to subsequent seismic activity.These variations are possibly due to the influence of energy emanating from inside the Earth's layers.Consequently, the first assumption upon which this research is based is apparently fulfilled, as monitoring the conductivity in areas with high seismic incidence and frequency indirectly provides a possible evidence of the presence of an electric and electromagnetic potential in the environment, energy that, as stated in the second assumption, could have interfered with the normal operation of the Data Logger, changing the movement of the ions dissolved in the aquatic environment towards a more intense field, resulting in reading variation of conductivity.• Variations or atypical conductivity readings in existing water bodies located in areas with frequent seismic activity show a subtle connection with the subsequent potential occurrence of earthquakes (Figures 12-17).• Conductivity readings in areas with high seismicity not follow a regular pattern over time, they are changing in short periods of time, registering up to three variations over a period of seven hours (Figure 14 and Figure 17).• Conductivity monitoring in seismic areas could be a sustainable option as a preventive measure for anticipating earthquakes.This can be achieved by placing measuring instruments in those areas where earthquakes historically occur.

Figure 1 .
Figure 1.Study area location as regards the seismic regions of Mexico.The area has the highest number of earthquakes per km 2 , corroborated by processing 12757 seismic events (red points) in a geographic information system, occurred in the states of Guerrero and Oaxaca, Mexico, from 2006 to 2014 with a range of magnitude between 3.5 to 7.4 Mw.Additional Mapping information was obtained of National Commission for Knowledge and Use of Biodiversity (CONABIO: http://www.conabio.gob.mx).

Figure 4 .
Figure 4. Conductivity pattern behavior in the "Los Manantiales" site.Its monitoring was suspended for reasons mentioned above.The average values of the quality of water obtained with HI 9828 were: Temperature (˚C) 29.18 pH 6.57; 0.97 ppm Dissolved Oxygen and 0.91 salinity.

Figure 5 .
Figure 5. Conductivity pattern behavior in the "Cistern" site, period November 04 to 11.Total seismic events recorded during monitoring: 18.The average values of the quality of water obtained with the HI 9828 were: T (˚C) 24.35 pH 9.31, 5.51 ppm OD and 0.16 salinity.

Figure 6 .
Figure 6.Conductivity pattern behavior in the "Cistern" site, period November 15 to 22.Total seismic events recorded during monitoring: 24.The average values of the quality of water obtained with the HI 9828 were: T (˚C) 24.35 pH 9.31, 5.51 ppm OD and 0.16 salinity.

Figure 7 .
Figure 7. Conductivity pattern behavior in the "Cistern" site, period December 06 to 12. Total seismic events recorded during monitoring: 12.The average values of the quality of water obtained with the HI 9828 were: T (˚C) 24.35 pH 9.31, 5.51 ppm OD and 0.16 salinity.

Figure 8 .
Figure 8. Conductivity pattern behavior in the "Cistern" site, period December 13 to 20.Total seismic events recorded during monitoring: 14.The average values of the quality of water obtained with the HI 9828 were: T (˚C) 24.35 pH 9.31, 5.51 ppm OD and 0.16 salinity.

Figure 9 .
Figure 9. Conductivity pattern behavior in the "Cistern" site, period December 22 to 27.Total seismic events recorded during monitoring: 8.The average values of the quality of water obtained with the HI 9828 were: T (˚C) 24.35 pH 9.31, 5.51 ppm OD and 0.16 salinity.

Figure 10 .
Figure 10.Conductivity behavior pattern of the "Teoixtla" site, period September 29 to October 04.Total seismic events recorded during monitoring: six.The average values of the quality of water obtained with the HI 9828 were T (˚C) 22.51 pH 7.18; 1.49 ppm OD and 0.28 salinity.

Figure 11 .
Figure 11.Conductivity behavior pattern of the "Teoixtla" site, period October 06 to 11.Total seismic events recorded during monitoring: 11.During this period a visible atypical variation of conductivity occurred on October 8.The average values of the quality of water obtained with the HI 9828 were T (˚C) 22.51 pH 7.18; 1.49 ppm OD and 0.28 salinity.

Table 1 .
Monitoring periods of selected sites.

Table 2 .
Quantification of seismic events used during the study at each site.

Table 3 .
General characteristics of most notably seismic events in the "Cistern" site.

Table 4 .
General characteristics of most notably seismic events in the site "Teoixtla".