An automatic system for continuous monitoring of CO2, H2S, SO2and meteorological parameters in the atmosphere of volcanic areas
© The Royal Society of Chemistry and the Division of Geochemistry of the American Chemical Society 2001
Received: 25 May 2001
Accepted: 13 July 2001
Published: 25 July 2001
An automatic system for the continuous monitoring of CO2, H2S, SO2 and meteorological parameters in atmosphere has been developed. The system has been tested in the laboratory in order to verify the stability and reliability of each sensor and of the whole system. A field test for a period of one month, at the Solfatara of Pozzuoli has also been carried out. The acquired data during the field test reveal a correlation between the wind speed and the concentrations of CO2, H2S, and SO2 in the atmosphere. With a wind speed of over 4 ms-1 the concentration of the three gases reached constant background values of 600 ppm for CO2 and about 2 ppm vol. for H2S and SO2. The different ratios of H2S/SO2 measured in the fumaroles (~100) and in the atmosphere (1–0.1) clearly indicate that H2S is oxidized to SO2 during the transport.
Active volcanic areas are generally indicated by different kinds of degassing, such as fumaroles, mofettes and diffuse soil degassing. The intensity of degassing, as well as the composition of the gaseous manifestations, are strictly related to the state of activity of each volcano. In the last years, several studies carried out in different volcanic areas clearly indicate that the presence of gas hazard is quite recurrent and that it is essentially controlled by the strength of the source, the morphology of the area, and meteorological parameters. [1–5]
According to the composition of volcanic gases, the main gaseous constituents that could be responsible for the existence of gas hazard are essentially CO2, H2S, and SO2.
The last two gases are normally present in areas close to fumaroles and the ratio H2S/SO2 in fumarolic gases can change in a wide range, being essentially controlled by the temperature and oxygen fugacity values of fumaroles (at higher temperature and fo2, SO2 is more abundant while H2S is the dominant species at lower temperatures and fo2). The CO2 is always present in all gaseous manifestations such as fumaroles, mofettes, diffuse soil degassing and it is the most abundant species of the incondensable gases.
It is then necessary to have the possibility of monitoring the content of these three gases in the atmosphere in order to control the gas hazard in a given area. The main objective of this work was to develop an automatic apparatus for the continuous and remote control of the concentrations of the aforementioned gases in the atmosphere, as well as the meteorological parameters (rain, wind speed and direction, atmospheric pressure and temperature).
Automatic station for continuous monitoring of CO2, SO2, H2S and meteorological parameters
Characteristics of the system
The system can essentially be divided into three parts:
1. Data logger (DL)
The DL used has been projected and assembled in the I.N.G.V. electronic laboratory. The main purpose to develop a new data logger was essentially due to the very corrosive environment normally present in volcanic areas.
The DL has 15 channels for analog inputs and each one may be connected to one sensor to read what follows: 0–2 V; 0–20 mA; K V, where K is a fixed constant for all the channels. In this way, the sensors normally used to monitor geochemical parameters (thermocouple, thermoresistance, electrochemical sensors, spectrophotometer IR, etc.) can be connected directly to the analog input of the data logger without any use of other transducers.
The DL is powered by a 12 DCV (direct current voltage) battery, and the battery is recharged by two solar panels. The total current consumption (electronic and sensors) is 600 mA. It takes a few minutes before reading some geochemical parameters; this depends on the activation of external devices like electrovalves, pumps, etc. It should be necessary to pilot it with a small computer, to run all these operations with a normal DL. In our DL, all these operations are temporised, only once at the boot time of the data logger and then, it repeats automatically the prefixed sequence, in all successive readings.
The data logger has 16 channels for digital output, ON/OFF type, and 8 digital input channels. Each one of them can send an alarm message.
In the initial boot of the data logger, for each analog channel input, some parameters are memorised as:
(i) Interval between successive readings.
(ii) Minimum and maximum alarm value of the parameter.
(iii) Pre-set to switch on and off a digital output channel at a prefixed time before the analog reading take place.
(iv) Data logger with a memory for more than 6000 readings.
(v) Maximum power consumption is when the data are discharged via radio.
The acquired data, stored in the DL memory, can be remotely discharged through a radio link. If some parameters reach values outside the fixed maximum and minimum, a window will automatically send an alarm message.
This possibility is a good improvement because it is possible to discharge the data during the day, when the batteries are recharged by the solar panels. This configuration is a good compromise between electric power saving, data archive for research purposes and their immediate disposability for geo-chemical surveillance of volcanic activity.
Each remote station is equipped with sensors for: CO2, SO2, H2S and meteorological parameters.
The chemical sensors are manufactured by LSI S.p.A. (Laboratori Strumentazione Industriale).
The electronic weather monitor (temperature internal and external, relative humidity, barometric pressure, wind speed and direction, rainfall) is a Wm-918, furnished by the Huger Electronics GmbH.
3. Transmission system
The acquired data can be transmitted through direct computer interfacing using RS232, radio modem and cellular modem connection.
In the same figure a calibration signal of 1000 mV has also been plotted, it has been divided by ten in order to plot all signals in the same figure.
Very constant values of this last parameter, (1000 ± 2 mV) indicate the good performance of the overall electronics. In other words, the potential error introduced by the electronics of the system is ± 0.2% of the signal.
The concentration values of SO2 and H2S have been multiplied by ten in order to plot all signals in the same figure. The output of both sensors was stable, exhibiting variations in the range ± 0.1 ppm. The concentration values of CO2 in Fig. 2 have been divided by ten. All recorded values, corrected for the zero shift, are in the range 350–700 ppm. The peaks have been recorded during daytime when people were working in the lab. Data acquired from 23rd December to the end of the month (holiday period) exhibit CO2 concentration values in the range 350–450 ppm.
Temperature changes of a few degrees have been observed in both sensors (internal and external to the box of the station). Of course, the internal sensor always showed higher temperature values than the external one. The same pattern can be observed for what concerns internal and external relative humidity. Naturally, in the last case, the range of variations is wider.
Finally, as expected, small changes of a few millibars of pressure have been recorded. All the tested sensors show a very good long-term stability with any drifts of the signal.
The choice of the location of the monitoring site was a consequence of the location of the main fumaroles that are chiefly located on the eastern side of the Solfatara.
The station worked for about one month (18th June–12th July 1999) without any technical problems. The atmospheric gases were pumped through a Teflon tube with the open external side placed at 1 m from the ground. The pump outlets were connected to the sensors in series.
The selected frequency of measurements was 1 h. The wind speed during the period of observation was in the range 0–5 m s-1 and the dominant direction was from N–NE (220°).
strength and location of the source;
meteorological parameters and morphology of the area;
reactions during the transport.
During the period of observation we assume that the strength and the location of the source have been constant, as well as the morphology of the area. The observed variations could mainly be related to the other factors such as meteorological parameters and chemical reactions during the transport. Among the investigated species CO2 can be considered, from a chemical point of view, an inert gas while H2S and SO2 can initiate chemical reaction during the transport. In this context we will first analyse the behaviour of CO2 and that of the two other species.
With the wind speed above 4 ms-1, the CO2 concentration reached a value close to 600 ppm. This value seems to be the background value of the CO2 content in the atmosphere of the Solfatara.
H2S and SO2variations
On the other hand, the abundance of H2S and SO2 in the atmosphere is highly dependent on the strength of the source as well as the chemical reactions during the transport. The H2S content in the source gases (fumaroles) was 0.05% vol. The SO2 content was below the detection limit of the analysed fumarolic gases (5 ppm by gas chromatography), the H2S/SO2 ratio being higher than 100. This ratio is completely different from the ratio measured in the atmosphere that changed between 1 and 0.1, it is 2–3 orders of magnitude lower than that of the source gases. The explanation for this observation must take into account the reaction during the transport.
The most probable reactions that produce SO2 starting from H2S are the following:
H2S + O3 = SO2 + H2O
H2S + O = OH + HS
H2S + OH = HS + H2O
These reactions are completed in the following way:
HS + O2 = SO + OH
HS + O = SO + H
SO + O = SO2
SO + O2 = SO2 + O
SO + O3 = SO2 + O2
These mechanisms depend on the temperature, humidity, aerosol, particulate, insulation, and are characterized by rather different kinetics. It is not possible to clarify all possible oxidation mechanisms than can take place inside the Solfatara with the acquired data, even though the variability of the H2S/SO2 ratio (from 100 in the fumaroles to 1 or 0.1 in the atmosphere), gives evidence of the H2S oxidation.
Specific characteristics for the H 2 S electrochemical sensor
Range of measurement
± 0.3% signal °C-1
Specific characteristics for the SO 2 electrochemical sensor
BSO 1 1 1
Range of measurement
± 0.1% signal °C-1
Specific characteristics for the CO 2 sensor. The sensing element is an infrared spectrophotometer
Range of measurement
0.1% full scale °C-1
Specification of the electronic weather station
0 to 50 °C
± 1 °C
-40 to 60 °C
± 2 °C
0.1 °C typical
10 to 97% RH
Accuracy (at temp, range 15 to 40 °C)
± 5% RH
795 to 1050 mbar
Accuracy (at temp, range 0 to 50 °C)
± 7 mbar
0 to 56 m s-1
Accuracy (at temp, range -20 to 60 °C)
(1) range 2 to 10 ms-1
(2) range 10 to 56 m s-1
0.2 m s-1
0 to 359°
Daily and cumulative measuring range
0 to 9999 mm
Rainfall rate measuring range
0 to 998 mm h-1
Specific characteristics for the different system
Direct computer connection
from 600 to 9600 bps
600 or 1200 bps
VHP or UHF
Depending on the company
- Badalamenti B, Gurrieri S, Hauser S, Parello F, Valenza M: Rend Soc Ital Mineral Petrol, M Carapezza Memorial Volume. 1988, 43 (4): 893-Google Scholar
- Badalamenti B, Gurrieri S, Hauser S, Tonani F, Valenza M: Rendiconti SIMP. 1984, 39 (2): 367-Google Scholar
- Chiodini G, Frondini F, Raco B: Bull Volcanol. 1996, 58: 41-10.1007/s004450050124.View ArticleGoogle Scholar
- Gerlach TM: Eos Trans. 1991, 72: 249-Google Scholar
- Graziani G, Martilli A, Pareschi MT, Valenza M: J Volcanol Res. 1991, 75: 283-10.1016/S0377-0273(96)00067-4.View ArticleGoogle Scholar
- Carapezza M, Nuccio PM, Valenza M: Bull Volcanol. 1981, 44: 547-View ArticleGoogle Scholar
- Carapezza M, Gurrieri S, Nuccio M, Valenza M: Bull Volcanol. 1984, 47 (2): 287-10.1007/BF01961559.View ArticleGoogle Scholar
- Badalamenti B, Carapezza M, Scalzo A: Gas hazard assessment in a densely inhabited zone near Rome (Cava del Selci, Alban Hills). in preparation
- Cox RA, Sandalls FS: Atmos Environ. 1974, 8: 1269-10.1016/0004-6981(74)90006-7.View ArticleGoogle Scholar