The scientific case.

The study of the snow accumulation at the Earth's cold regions, glaciers and polar ice-caps, is of great importance in fields like glaciology, meteorology and climatology. One way to measure snow deposition history is by visual stratigraphy the other is by identifying layers of chemicals or particles that vary periodically with seasons or are associated with historical events such as volcanic eruptions and thermonuclear tests or accidents. Because of their well-known dates, these events serve as characteristic reference levels in snow. Artificial and natural radionuclides present in the ice sheets of the cold regions are currently used, in glaciological studies, to determine the average rate of snow accumulation over the last 40 years and have also been widely used as atmosferic tracers to understand the transport and circulation of continental, tropospheric and stratosferic air masses. Chemical analysis has given the possibility to determine, as shown in Fig. 1, the evolution of the environmental pollution produced by human activity ¹ while, as shown in Fig. 2, particle analysis in ice samples, as a function of depth, has been used to determine absolute time markers ² since the peaks found were linked to nuclear explosions in fifties and sixties and to the Chernobyl accident ³ (1986).

Fig. 1 Evolution of the environmental pollution produced by human activity.

Fig. 2 Beta activity vs. depth in ice core (from ref. [2]).

Chemical and particle analysis usually requires samples from snowpits or ice cores to be gathered and returned to the laboratory. Since these samples are often taken from places that are remote, handling and shipping the samples can be very difficult since some places are of difficult access. This is particularly true in the higher parts of mountain regions that are very sensitive to small amounts of radioactive contaminants and offer the unique opportunity to detect in advance small changes in the atmospheric composition. Therefore, it would be useful to have a way to determine, in the field, one or more absolute time markers in the snow ⁴ . This would allow the selection of samples for the specific time intervals of interest and could in some case eliminate the need for sample retrieval.
It has been found3,4 that in-situ measurements of 137Cs g -rays (half-life = 30.17 yr) from fallout can be used as an absolute time marker. This implies lowering a detector down in bore-holes or along trench profiles in the snow to measure the characteristic 662 keV g -rays emitted by Cs-137. Among the main fission product from nuclear tests or accidents, Cs-137 is the gamma -emitter with the longest half-life.
The g -ray detection method is of great interest for the Italian Glaciological Community involved in national and international glaciological studies some of which are also included in the activities of the Ev-K2-CNR Project and of the new Istituto Nazionale per la Ricerca Scientifica e Tecnologica sulla Montagna (INRM).

Main interests of this kind of research.

The information that can be collected by this kind research is very useful in order to evaluate local and global change processes, including the radio-chemical pollution, induced by human activity.
International scientific communities are promoting studies in remote areas and at high altitude to monitor and control :

• local and global climate changes;
• environmental pollution;
• reduction of the high altitude glaciers.

Future perspectives.

The detector will be used in glaciological studies in snowfields and high altitude glaciers (Gran Sasso, Mt. Rosa, Mt. Bianco, Greenland, Antarctic and Himalaya). In particular it will be used during the " Roma 8000 " expedition at the Cho-Oyu mountain in Tibet (15August-15 October 2000). This expedition related to the Italian activities for the year 2002 proclaimed by the UN to be the " International Year of the Mountains ", is also sponsored by INFN and will represent for the detector the first very high altitude test.

Technical aspects.

A . Sensitivity, resolution and efficiency .

A gamma-ray spectrometer based on a NaI(Tl) scintillator crystal coupled with a photomultiplier, has a high efficiency in the Cs-137 gamma-ray region and enough resolution to discriminate the Bi-214 peak at 609keV which is of natural origin. In order to achieve a very low minimum detectable signal, a great scintillator volume is needed, and since the bore-holes in snow have a minimum diameter of 100 mm, a cylindrical scintillator crystal with 76 mm diameter and length of about 100 mm was selected.
The method to measure 'in situ' radioactive fallout, monitoring the emission at 662 keV of Cs-137 by using a NaI(Tl) scintillator detector coupled with a photomultiplier tube and a multichannel analyser was proposed in 1981, by Pinglot and Pourchet3. Their measurements have shown that the gamma activity can be detected 'in situ' and that it is strongly correlated with the radioactivity measured in samples. The limit of their system was the portability, being the weight of the whole detector system of about 250 kg !
In 1994, Dunphy, Dibb and Chupp4 proposed a lighter portable system based on a NaI(Tl) scintillator detector computer controlled. Their measurements were done near a permanent base by using a standard AC power source which is not usually available - or portable - in some remote areas, like high altitude glaciers. For this reason GEDI will develop a completely portable instrument that can be easily carried and handled by a small team without any external support and the scintillator, the photomultiplier, the high-voltage divider and the preamplifier will be assembled and housed in a thin stainless steel water-resistant cylinder in order to be able to work in harsh environmental conditions.

B . Power supply.

As power source a mixed system composed by batteries and photovoltaic cells will be used. Considering that the power required by the computer and the detector is lower than 100 W and that 100 W is the evaluation of the power needed to warm up the detector when working in very low temperature conditions, about 2 m2 of Si photovoltaic cells are necessary. This power is assumed to be the maximum consumption limit of our system. The Si photovoltaic cells will be chosen considering their weight and portability. As shown in Fig. 3 also the possibility to use wind generators, was taken into account considering different environmental conditions: a wind generator of 6 kg weight - working with a wind speed of about 10 m/s - can generate a power of about 300 W.

C . Low temperature conditions.

Fig. 4. Responces of NaI(Tl)and of other scintillators as a function of temperature.

The detector can be exposed to temperatures as low as -50°C. Since the temperature gradient can damage the detector and affects the detector response (Fig. 4) changing the position and the width of the detected peak, the stainless steel cylinder will be wrapped with a flexible heater controlled by a thermostat. To protect from water and mechanical shocks, the all system (detector, heaters and thermostat) is housed in an acrylic tube. The temperature is monitored (inside and outside the housing tube and in the environment) by using semiconductor temperature sensors.

D . Portability. All the detector system must be carried with backpacks by a team of two or three persons. So all its parts must be accurately selected specially as a function of weight in order to obtain an overall weight not higher than 15-20 kg. For this reason the use of a portable generator as power supply at very high altitude was excluded. In some preliminary tests and measurements on the Calderone Glacier (2700 m) a portable generator of about 13 kg was used as shown in Fig.5.

Fig. 5. Portable power generator in use on the Calderone Glacier.

E . Data acquisition. Data acquisition and temperature controls will be performed by using a Panasonic (Toughbook CF 27) portable computer (Fig. 6) - protected against water, dust and mechanical shocks - and a National Instruments DAQ Card (AI-16E-4) that allows the input/output of analog and digital signals. A stretching circuit has been developed to match the preamplifier output signal with the input requirements of the card. All the signals will be processed by means of LabVIEW (5.1) software able to reproduce a multichannel analyser system.

Fig. 6. Panasonic (Toughbook CF 27) portable computer during data acquisition.

Preliminary tests and first results. Some preliminary tests have been done on the field, on the Calderone glacier (Fig. 7), by using a commercial portable detector. The tests have shown that even in mild climate conditions it is necessary to work with a suitable instrument since a commercial one does not satisfy the requirements for 'in situ' measurements. In June 1999 the following field operations were performed: general measurements of thickness of snow and choice of the best site; digging of a trench in the whole thickness of the whole snow layer, deposed above the glacier body and the superficial debris; stratigraphy of the snow thickness and, in the first sampling, characterisation of physical and mechanical properties according to the international classification of snow; sampling of the most representative layers of the trench. Two samples of snow, belonging to the layers 2 and 9, were gathered to perform laboratory measurements. The non-filtered samples were put in Marinelli's bekers (500 cm3 capacity) and, in order to reduce the environmental radioactivity contribution, were well shielded by a 10 cm lead layer and two layers of copper and cadmium, of 1 mm each. The acquisition time was about 170000 seconds. The measurements on the snow samples were performed by using a low-background laboratory gamma spectrometer composed by an HPGe - high purity Germanium - solid state gamma ray detector (PGT) with 16% efficiency and a resolution (FWHM) of 1.9 keV/1.33 MeV. As shown in Fig. 8 the spectra show clearly the presence of Cs-137 gamma peak, at the energy of 662 keV. The other higher peak at 609 keV is due to the natural isotope Bi-214. Strong quantities of Cs-137 were sent in the atmosphere during thermonuclear tests (for example in the years 1954 and 1962) and nuclear accidents, like Chernobyl in 1986. Traces of this last event are still present in the environment and as shown by these measurements small quantities of Cs-137 are still present in the atmosphere and can be detected in snow layers associated to single atmospheric events. The results achieved in these preliminary measurements 6 and the GEDI experiment 7 were presented at the International Symposium on 'Global Changes and Protected Areas' that was held in L'Aquila in September 1999.

Fig. 7. The whole apparatus used during some preliminary 'in situ' tests on the Calderone Glacier.

Fig. 8. The gamma-ray spectra of the 1999 sample (layer 2). The peaks due to 662 keV Cs-137 and 609 Bi-214 gamma-rays are indicated.

References.

1. P.A.Mayewski et al., Nature 346, (1990) 554-556
2. J.B.Barnola et al., Nature 329, (1987) 408-414
3. J.E.Dibb et al. Nature, 345, (1990) 25
4. P.P.Dunphy, J.E.Dibb, E.L.Chupp, Nucl. Instr. Meth A 353 (1994) 482-485
5. J.F.Pinglot and M. Pourchet, In: Methods of low-level counting and spectrometry, IAEA, Vien (1981), pp 161-172
6. E. Bernieri et al., “Cs-137 gamma peak detection in snow layers on Calderone glacier”, Proc. Symposiun on Global Change and Protected Areas, L'Aquila 8-13 september 1999, to be published
7. E. Bernieri et al., “Gamma-ray spectrometer for in situ measurement on glacier and snowfields”, Proc. Symposiun on Global Change and Protected Areas, L'Aquila 8-13 september 1999, to be published

Appendix : Difficulties in performing 'in situ' measurements at high altitude: