Oxford University Cave Club

Water Analyses and Hydrology

There are two primary factors which influence the formation of caves: the nature of the rock formations and the physical and chemical characteristics of the ground water. Of these the former is the province of the geologist and the geomorphologist the latter is the concern of the chemist. This article describes the chemical measurements performed on the expedition.

It was thought to be of interest to undertake a limited survey of the natural waters of the Picos area. In practice this was limited almost exclusively to the Trema spring near the refugio and the rising at the Vega de la Cueva. The Trema spring emerges through a permanent piped springhead and supplies water for the refugio; it gives rise to a tiny stream which after a short course drains into Lago Enol. The Vega de la Cueva rising emerges at the side of the bed of a stream, the upper course of which was dry during our stay, perhaps due to the exceptionally dry conditions. This stream runs from the opposite side of the watershed on the saddle near the refugio and has carved a fairly deep valley, eventually joining the Rio Dobra. Some comparisons between the characteristics of these two risings are made later.

The altitude of the main spring was 1100 metres and that of the rising at the Vega de la Cueva 1050 metres; they were separated by a distance of 2/3 Km.

Because it was not possible to transport sufficient chemicals and apparatus to Spain to perform all the analyses which would have been of interest it was decided to concentrate on the simplest and most important. Even these required a large amount of apparatus. The expedition went to Spain equipped to measure water temperature and rate of flow, the hydrogen ion concentration (pH) and the total calcium and magnesium content.

Experimental methods

Samples of water were taken once daily at convenient times. Temperature and flow rate were measured simultaneously and a sample removed to the refugio for analysis. Temperature was measured during the earlier part of the period by using a photographic alcohol thermometer housed in a metal case. Readings were taken in degrees Fahrenheit to an accuracy of +/- ¼ °F.; these have been converted to °C. and reported to the nearest 0.5 °C.

In spite of the robust construction of these thermometers both of those available were eventually broken and replaced by a small alcohol thermometer calibrated in °F. bought locally in Cangas de Onis. Although this thermometer survived for a long time it did not return to England and so it is not possible to estimate accurately its reliability. This change took effect from August 28th. The thermometer was placed in the direct flow of the water out of direct sunlight for at least one minute before the reading was taken. Rate of flow was the simplest measurement to perform on the Trema spring, as the water issued from its pipe in a single stream which could be caught in a bucket. The time required to fill to the gallon mark was recorded by means of a stop-clock. In Table 1 this has been converted to gallons per minute.

In the case of the rising at Vega de la Cueva several minor springs joined the stream and it was necessary to constrain as much of the flow as possible to run over a plastic sheet which was arranged to direct it into a bucket. An estimate was made of the water flow which was not recorded in this way.

The standard laboratory pH meter working from a mains electricity supply is not suitable for use on expeditions, and battery-powered portable instruments were not readily available at the time. Unfortunately no simple method of pH measurement approaches the accuracy of a pH meter which can be read to 0.01 of a pH unit. When a meter is not available, the most convenient method of estimating pH values utilises the colour changes of indicator dyes. Under carefully controlled conditions this is capable of providing an answer to within 0.1 of a pH unit.

An exceptionally simple means of obtaining standard conditions in the field is provided by the B.D.H. Lovibond Nessleriser (Plate 6). This consists of a cabinet containing two matched tubes which are filled to a graduated mark with the water under test. To the right-hand tube 0.2 ml. of an indicator dye is added from a dropper pipette. The tubes are illuminated from below by a diffused light reflector and observed from above. The left-hand is covered by a disc which contains the colours corresponding to the pH-range of indicator in steps of 0.2 pH units: estimations are possible to 0.1 unit The disc is rotated until a colour match is obtained between the two tubes. A separate comparison disc is supplied for each indicator dye.

The following indicator dyes were used:

Where possible it is advisable to check readings by using two indicators having overlapping ranges because readings at the extreme ends of the range are not reliable. A useful addition to the above indicators would be phenol red of pH range 6. 8 to 8.4 (transitional).

The Nessleriser was used only for the determination of pH and Hazen colour values. Methods are available to enable estimations to be performed of various impurities in water by means of the same instrument. Full details of all applications may be obtained from British Drug Houses Ltd, Poole, Dorset.

Hardness determination

The methods of analysis for calcium and magnesium were chosen so as to provide the maximum of information with the minimum of apparatus. This plan was upset by two unfortunate occurrences. Firstly it was found on arrival in Spain that the two burettes required for titrations had not arrived. Accordingly as many samples as possible were kept to bring back to England for analysis. Secondly the total supply of one indicator was lost on the return journey when its container broke. Thus the results of only one series of titrations can be given in this report.

If large-scale chemical analyses are to be performed in the field with no access to laboratory facilities there are a number of obstacles to be overcome. Probably the most serious is the provision of a sufficient quantity of distilled water. It was possible for the expedition to take 30 litres of water in polythene carboys, but if transport cannot be provided for such large quantities it would be necessary to distil water in situ. This would be at the least an arduous task. It was found that a good method of transporting chemicals for standard solutions was to weigh out aliquots of the correct size before departure. These may then be sealed inside plastic containers to be made up as required. This eliminates all problems of ageing of solutions, breakage, evaporation etc., but it increases the demand on distilled water.

The solutions which were transported in plastic bottles suffered no losses with one serious exception. However, no special precautions were taken to ensure their safety. In contrast all the glass equipment, pipettes, flasks etc. travelled without loss but needed to be packed very carefully. It is impossible at present to use plastic apparatus exclusively, but the more plastic that can be taken the better. However, from the experiences gained from a not very ambitious analytical programme the conclusion might be drawn that it is better not to try to take too much equipment. The space taken by analytical equipment is more profitably filled with polythene bottles to bring back samples for analysis in the laboratory at home. In this way much more information may be obtained.

Temperatures and flow times must be measured on the spot, but they need very little apparatus. Unfortunately pH must be measured as soon as possible, as it can change appreciably on standing. The simplest way to measure this, in the absence of a portable battery-powered pH meter is by close-range indicator papers as supplied by Johnsons. The Nessleriser was used, but as the glasses which hold the samples are very fragile, a number of spares were taken.

A valuable addition to any report of water analyses is an analysis of rock specimens collected in the same area. Limestone is relatively easy to analyse, consisting of over 90% calcite or dolomite, with the balance made up by quartz, felspars, clay minerals, iron ores and a few other minerals.

The records of the Trema and Vega de la Cueva springs given in Tables 1 and 2 may be compared with interest. The prevailing weather conditions are also tabulated in Table 2. The records cover two periods of rainy weather, 18 - 20 August and 1 - 9 September. Both springs react within 24 hours to rain, as can be seen in the Trema records of 21st August, and for both springs on 6th and 7th September. The Trema spring, however, rapidly rises and subsides, whereas changes for the Vega de la Cueva rising are much more gradual. This may be due to the fact that the Trema spring probably drains a limited area under the hill behind the refugio, while a much greater area is involved for the Vega de la Cueva rising, which is shown by the large volume of water at this rising and the rate of decline of the flow rate during the prevailing dry weather. The Vega de la Cueva rising showed a 90% reduction of flow rate in 24 days, while the Trema spring showed only a 60% reduction over the same period. The Vega de la Cueva rising is more alkaline than the Trema spring (pH 7.7 mean compared with pH 7.4 mean), further supporting the hypothesis that the Trema spring has the more limited gathering area. Temperature of the two springs in general show no relation to rainfall, while pH is constant for both within the limits of experimental accuracy, with the exception that the Vega de la Cueva rising seems to show fluctuations during the rainy period 1-9 Sept.

The most probable source for the water of the Vega de la Cueva rising was always thought to be the sink designated S1 in the 'Hidden Valley' of Las Reblagas (see Figure 1). The stream here is of similar size to that issuing from the Vega de la Cueva. rising (designated R1), but attempts to establish the connection with Fluorescein and with Rhodamine B on separate occasions during 1961 failed. The stream at Las Reblagas itself rises from the hill slopes at the northeastern corner of the completely enclosed valley (polje), and has been proved to connect with the sink at the marshy southern end of Lago de la Ercina (S4 to R2). S1 has now been proved to connect with rising R12 in the Vega de la Cueva. The colouring matter took about five weeks to travel half a mile in 1961. This occurred after the expedition had left the area, but we were informed by the local inhabitants. Also a possible source for R1 may be the active stream found at the inner end of C15 or water from the Pozo Palomeru (P1) system of passages.

The substances used for hydrological research on any expedition must be a matter of personal preference now that a variety of chemicals are available for the purpose. We used Fluorescein and Rhodamine B, the latter by using the method developed by members of the Bradford Pothole Club (5); this requires the provision of cotton hanks, tannic acid and potassium antimonyl tartrate for mordanting, E.D.T.A. (sodium salt) to remove the dark-blue discoloration caused by iron salts often present at the rising, and means to hold the cotton hanks immersed at the rising. It was found convenient to prepare the hanks of cotton during the time of the stay in the Picos, but others may find it more convenient to prepare a supply of hanks for detection purposes before leaving Britain. The amount of dye to be used is best determined by experiment on a number of known interconnected sinks and resurgences in Britain, preferably chosen to include a large range of flow rates. It is better to use a concentrated solution of dye rather than the powder - both Fluorescein and Rhodamine B can stain skin or clothing badly, and can cause one to come out in green or red spots if particles are carried by the wind. Fluorescein in quantities not great enough to colour the rising visibly may be detected with adsorbent charcoal contained in bags made of wide roller-bandage, and this is perhaps the most convenient method for a remote area (6).

The natives of the area must of course always be considered when using colouring agents in water. Many streams in remote areas are the only water supply for animals and humans, and occasionally a rising may be at a religious shrine, as at Covadonga in the Picos de Europa.

The drainage of the whole area of the Picos de Europa is most interesting. The hydrological work of the expedition was hampered by the long spell of dry weather, but samples of water were obtained from all the investigated sinks and risings. Tables 3 and 4 give a summary of sinks and risings studied by the expedition, using numerical designations, and should be studied in conjunction with Figure 1. The drainage of the two lakes in the area deserves much further study.

Lago de Enol receives only a tiny trickle of surface water from R5, but the outflow is substantial, and is used partly for dressing ore at the Minas de Buffarera. It seems likely, therefore, that there is an underground water source entering this lake. Lago de la Ercina seems to have approximately equal rates of inflow and outflow, but the lake does not have an outlet down the valley at its northern end, at least under summer conditions. Instead the outflow is through a small cave (C4) at the southern end, and by underground connection to the polje Las Reblagas, finally sinking again.

The impressive polje Vega de Comeya, containing the processing plant for the Minas de Buffarera, and almost entirely ringed by 200 foot cliffs, contains a very strong stream from a number of risings, one of which (R9) lies at the foot of the cliffs.

This stream sinks at four places (S8 - S11) at the western side of the polje, but attempts to connect these with the waters of the Rio Arganeo, Agua de Santa Marina or La Guanaz all failed. The rising must be of large volume and awaits further discovery.

It is hoped that this overall impression of the drainage of the area will form a basis upon which further plans for speleological research can be based.

M. Austin, Shenfield, Essex. April 1964; J.D. Wilcock, Kidsgrove, Staffordshire. June 1964.

References

1. Journal of the American Chemical Society, 75, 4196: 1953; Journal of the American Waterworks Association, 42, 33 and 49: 1950.
2. Radex Rundschau, 333: 1955.
3. Analytical Chemistry, 29, 264: 1957.
4. "Approved Methods for the Physical and Chemical Examination of Water", Institution of Water Engineers, 1960.
5. Bulletin of the Bradford Pothole Club, Vol. 3, No. 3 11: Nov. 1960.
6. N.S.S. Bulletin, Vol. 21, Part 2: July 1959.

Table 1: Trema Spring Records

Table 2: Records of the rising in the Vega de la Cueva

Explanation of weather symbols:

Preceding weather conditions are separated from prevailing weather conditions by an oblique stroke: /

Table 3: Record of Sinks Investigated by OUENS61

Table 4: Record of Risings Investigated by OUENS61