The utility of stable isotopes as tracers of the water molecule has a long pedigree. The study reported here is part of an attempt to establish a comprehensive isotopic 'baseline' for the British Isles as background data for a range of applications. Part 1 of this study (Darling and Talbot, 2003) considered the isotopic composition of rainfall in Britain and Ireland. The present paper is concerned with the composition of surface waters and groundwater. In isotopic terms, surface waters (other than some upland streams) are poorly characterised in the British Isles; their potential variability has yet to be widely used as an aid in hydrological research. In what may be the first study of a major British river, a monthly isotopic record of the upper River Thames during 1998 was obtained. This shows high damping of the isotopic variation compared to that in rainfall over most of the year, though significant fluctuations were seen for the autumn months. Smaller rivers such as the Stour and Darent show a more subdued response to the balance between runoff and baseflow. The relationship between the isotopic composition of rainfall and groundwater is also considered. From a limited database, it appears that whereas Chalk groundwater is a representative mixture of weighted average annual rainfall, for Triassic sandstone groundwater there is a seasonal selection of rainfall biased towards isotopically-depleted winter recharge. This may be primarily the result of physical differences between the infiltration characteristics of rock types, though other factors (vegetation, glacial history) could be involved. In the main, however, groundwaters appear to be representative of bulk rainfall within an error band of 0.5‰ δ 18 O. Contour maps of the δ 18 O and δ 2 H content of recent groundwaters in the British Isles show a fundamental SW-NE depletion effect modified by topography. The range of measured values, while much smaller than those for rainfall, still covers some 4‰ for δ 18 O and 30‰ for δ 2 H. Over lowland areas the 'altitude effect' is of little significance, but in upland areas is consistent with a range of -0.2 to -0.3‰ per 100 m increase in altitude. Groundwaters dating from the late Pleistocene are usually modified in δ 18 O and δ 2 H owing to the effects of climate change on the isotopic composition of rainfall and thus of recharge. Contour maps of isotopic variability prior to 10 ka BP, based on the relatively limited information available from the British Isles, allow a first comparison between groundwaters now and at the end of the last Ice Age. The position of the British Isles in the context of the stable isotope systematics of NW Europe is reviewed briefly.

1-2 ka for the desert's southeastern margin is inferred from radiocarbon. These results reveal that some replenishment to the desert aquifer is occurring but at a rate much lower than previously suggested, which is relevant for water resources planning in this ecologically-sensitive area.

No abstract

An understanding of the hydrological cycle in stable isotopic terms requires the characterisation of rainfall. This paper reviews existing and new data for the British Isles. Rainfall at the Wallingford (Oxfordshire) collection station was collected daily from November 1979 to October 1980. Large variations in isotopic content were noted, sometimes from day to day. Winter rainfall was similar to summer in amount, and only slightly depleted isotopically. Amount and temperature correlations with δ 18 O were generally low, only the autumn and winter temperature relationships being significant. A 20-year monthly dataset from 1982 to 2001 for Wallingford gives the following regression: δ 2 H = 7.0δ 18 O + 1.2, a slope somewhat below the world meteoric line but consistent with the those from other long-term stations in NW Europe. The data showed uncorrelated maxima and minima for each year, but rather more consistent amount-weighted averages. Although there is only a small difference in gradient between summer and winter rainfall values, when plotted against the month of the year there are clear changes in the values of both isotopes, and the δ 2 H-δ 18 O relationship as demonstrated by the d-excess parameter. The isotope-amount correlation is low but significant, with summer months appearing to be well-correlated when considered in terms of month of the year. On this same seasonal basis temperature has a strong correlation throughout the year, giving a positive δ 18 O-temperature relationship of 0.25 ‰ per°C change. The Wallingford monthly record is compared with data from Keyworth (Nottinghamshire) and the Valentia station of the GNIP (IAEA-WMO Global Network for Isotopes in Precipitation) in SW Ireland. While not large, differences between the stations are broadly attributable to the balance between maritime and continental influences. Over the period September 1981 to August 1982 the maximum number of monthly collection stations was operating across the British Isles. While a comparison of the sites serves mostly to illustrate the variability of British weather in space and time, there is clear isotopic evidence for the predominance of frontal rainfall in winter and convective rainfall in summer. The effect of altitude on isotopic content was measured within a high-relief stream catchment in Scotland. The best correlations occurred during winter, when an average relationship of approximately -0.30 ‰ δ 18 O per 100 m increase in altitude was observed. It is well established that rainfall isotopic composition changes in response to alterations in climate. However these changes are difficult to detect isotopically in the short term, even when the changes are indexed, e.g. in the form of the NAO (North Atlantic Oscillation). The brief duration of rainfall isotope records is a further hindrance; for the British Isles proxies such as tree-ring cellulose may have some value in extending the record back.

No abstract

Knowledge of groundwater residence time is important in understanding key issues in the evolution of water quality, whether this occurs due to water-rock interaction or simply by mixing or contamination. The build-up in the atmosphere of the trace gases chlorofluorocarbons (CFCs) and sulphur hexafluoride (SF 6 ) from the middle of the last century offers a convenient way of dating waters up to ~60 yrs old. The gases are well-mixed in the atmosphere so their input functions are not area-specific as is the case with tritium. While any one of these trace gases can in principle provide a groundwater age, when two or more are measured on water samples the potential exists to distinguish between different modes of flow including piston flow, exponential flow and simple end-member mixing. As with all groundwater dating methods, caveats apply. Factors such as recharge temperature and elevation must be reasonably well-constrained. Primarily for SF 6 , the phenomenon of 'excess air' also requires consideration. Primarily for the CFCs, local sources of contamination need to be considered, as do redox conditions. For both SF 6 and the CFCs, the nature and thickness of the unsaturated zone need to be factored into residence time calculations. However, as an inexpensive dating method, the trace gases can be applied to a wide range of groundwater problems where traditional age indicators might once have been used more sparingly. Examples include tracing flowlines, detecting small modern inputs in 'old' waters, and pollution risk assessment. In the future, with the main CFCs are already declining in the atmosphere, new anthropogenic trace gases are likely to take their place.

One of the key uncertainties surrounding the impacts of climate change in Africa is the effect on the sustainability of rural water supplies. Many of these water supplies abstract from shallow groundwater (<50 m) and are the sole source of safe drinking water for rural populations. Analysis of existing rainfall and recharge studies suggests that climate change is unlikely to lead to widespread catastrophic failure of improved rural groundwater supplies. These require only 10 mm of recharge annually per year to support a hand pump, which should still be achievable for much of the continent, although up to 90 million people may be affected in marginal groundwater recharge areas (200-500 mm annual rainfall). Lessons learnt from groundwater source behaviour during recent droughts, substantiated by groundwater modelling, indicate that increased demand on dispersed water points, as shallow unimproved sources progressively fail, poses a much greater risk of individual source failure than regional resource depletion. Low yielding sources in poor aquifers are most at risk. Predicted increased rainfall intensity may also increase the risk of contamination of very shallow groundwater. Looking to the future, an increase in major groundwater-based irrigation systems, as food prices rise and surface water becomes more unreliable, may threaten long-term sustainability as competition for groundwater increases. To help prepare for increased climate variability, it is essential to understand the balance between water availability, access to water, and use/demand. In practice, this means increasing access to secure domestic water, understanding and mapping renewable and non-renewable groundwater resources, promoting small-scale irrigation and widening the scope of early warning systems and mapping to include access to water.Key words groundwater; climate; Africa; water supply; drought; agriculture Quel impact aura le changement climatique sur les approvisionnements ruraux en eaux souterraines en Afrique? Résumé Une des incertitudes principales concernant les impacts du changement climatique en Afrique est l'effet sur la durabilité des approvisionnements ruraux en eau. Nombre de ces approvisionnements prélèvent dans des nappes souterraines peu épaisses (<50m) et sont l'unique source sûre d'eau potable pour les populations rurales. Les analyses des études existantes de précipitations et de recharge suggèrent que le changement climatique est peu susceptible de conduire à une défaillance catastrophique généralisée des ressources rurales améliorées en eaux souterraines. Ces dernières ne nécessitent que 10 mm de recharge par an pour satisfaire une pompe manuelle, ce qui devrait être réalisable pour une grande partie du continent, bien que 90 millions de personnes puissent être affectés dans des zones marginales de recharge en eaux souterraines (200-500 mm de pluviosité annuelle). Les leçons apprises sur les comportements des ressources souterraines au cours des sécheresses récentes, renforcées par des modélisations hydrogéologi...

No abstract