Advertisement

Characterization of recharge mechanisms in a Precambrian basement aquifer in semi-arid south-west Niger

  • Maman Sani Abdou BabayeEmail author
  • Philippe Orban
  • Boureima Ousmane
  • Guillaume Favreau
  • Serge Brouyère
  • Alain Dassargues
Report
Part of the following topical collections:
  1. Determining groundwater sustainability from long-term piezometry in Sub-Saharan Africa

Abstract

In the central part of the semi-arid Dargol Basin of southwestern Niger, most of the groundwater resource is contained in the fractured aquifers of the Precambrian basement. The groundwater resource is poorly characterized and this study is the first attempt to better describe the recharge mechanisms and hydrogeochemical behaviour of the aquifers. Hydrogeochemical and piezometric methods were combined to determine changes in recharge rate and origin of groundwaters for the shallow weathered aquifer and the deep fissured/fractured aquifer. At the basin scale, the groundwater fluxes towards the Niger River are influenced mainly by topography, with no visual long-term trend in groundwater levels (1980–2009). The hydro-geochemical signature is dominated by the calcic-bicarbonate to magnesian (70%) type. It shows evolution from an open environment with CO2 and low mineralized water (granitoids, alterites) towards a more confined environment with more mineralized waters (schists). Stable water isotopes (δ18O, δ2H) analysis suggests two main groundwater recharge mechanisms: (1) direct recharge with nearly no post-rainfall fractionation signature and (2) indirect recharge from evaporated surface waters and/or stream-channel beds. Groundwater tritium content indicates that recharge is mostly recent, with an age less than 50 years (3H > 3 TU), with only 10% indicating low or even no recharge for the past decades. A median value of the groundwater renewal rate estimated from individual values of tritium is equivalent to 1.3% year−1, close to the one determined for groundwater samples dating to the early 1980s, thus indicating no measurable long-term change.

Keywords

Niger Fractured aquifers Hydrochemistry Environmental isotopes Groundwater recharge Sub-Saharan Africa 

Caractérisation des mécanismes de recharge d’un aquifère de socle Précambrien dans la zone semi-aride du sud-ouest du Niger

Résumé

Dans la partie centrale du bassin semi-aride du Dargol du sud-ouest du Niger, l’essentiel des ressources en eau souterraine est contenu dans les aquifères fissurés du socle Précambrien. La ressource en eau souterraine est peu caractérisée et cette étude est la première tentative pour mieux décrire les mécanismes de recharge et le comportement hydrogéochimique des aquifères. Une méthodologie combinant l’étude de la piézométrie et l’hydrogéochimie est élaborée pour déterminer les changements dans les taux de recharge et l’origine des eaux souterraines des nappes superficielles d’altérites et des aquifères profonds fissurés/fracturés. A l’échelle du bassin, les flux d’eau souterraine vers le Fleuve Niger sont principalement influencés par la topographie; aucune tendance à long terme dans les niveaux d’eau souterraine (1980–2009) n’est observée. La signature hydrogéochimique est dominée par un faciès bicarbonaté calcique à magnésien (70%). Elle montre une évolution depuis un milieu ouvert (granitoïdes, altérites) avec du CO2 et des eaux moins minéralisées, vers un milieu quasi-fermé (schistes) avec des eaux plus minéralisées. Les teneurs en isotopes stables (δ18O, δ2H) de la nappe aquifère indiquent deux principaux mécanismes de recharge: (1) une recharge directe par les eaux des pluies peu ou pas évaporées, et (2) une recharge indirecte par les eaux évaporées issues des lits des koris et des eaux de surface. Les teneurs en tritium des eaux souterraines indiquent que la recharge est récente, avec un âge inférieur à 50 ans (3H > 3 TU), avec seulement 10% des points indiquant une recharge faible ou absente lors des dernières décennies. Le taux de renouvellement médian estimé à partir des valeurs individuelles en tritium est d’environ 1.3% an−1; proche de la valeur déterminée avec des échantillons datant du début des années 1980 ce qui indique qu’aucun changement sur le long terme n’est mesurable.

Caracterización de los mecanismos de recarga en un acuífero precámbrico en el sudoeste semi-árido de Níger

Resumen

En la parte central de la cuenca semiárida de Dargol, en el sudoeste de Níger, la mayor parte del recurso de agua subterránea está contenido en los acuíferos fracturados del basamento precámbrico. El recurso subterráneo está pobremente caracterizado y este estudio es el primer intento de describir mejor los mecanismos de recarga y el comportamiento hidrogeoquímico de los acuíferos. Se combinaron métodos hidrogeoquímicos y piezométricos para determinar los cambios en la tasa de recarga y el origen del agua subterránea en el acuífero somero, relacionado a la meteorización y el acuífero fisurado/fracturado profundo. A la escala de la cuenca, los flujos de aguas subterráneas hacia el río Níger están influenciados principalmente por la topografía, sin que exista una tendencia de variación a largo plazo (1980–2009) de los niveles de agua subterránea. La firma hidrogeoquímica está dominada por el tipo de agua bicarbonatada cálcica a magnésica (70%). Muestra una evolución desde un ambiente libre con agua con CO2 y poco mineralizada (granitoides, alteritas) hacia un ambiente más confinado con aguas más mineralizadas (esquistos). El análisis de isótopos estables del agua (δ18O, δ2H) sugiere dos mecanismos principales de recarga del agua subterránea: (1) recarga directa casi sin señal de fraccionamiento posterior a la lluvia y (2) recarga indirecta de aguas superficiales evaporadas y/o de lechos de los canales de las corrientes. El contenido de tritio en el agua subterránea indica que la recarga es en su mayoría reciente, con una edad inferior a 50 años (3H > 3 TU), y solo el 10% indica una recarga baja o incluso nula durante las últimas décadas. El valor de la mediana de la tasa de renovación de agua subterránea estimada a partir de valores individuales de tritio es equivalente a 1.3% año−1, próximo al que se determinó para muestras de agua subterránea que datan de principios de los 80, indicando que no hay cambios mensurables a largo plazo.

尼日尔半干旱西南部前寒武基底含水层内补给机理特征描述

摘要

在尼日尔西南部半干旱的Dargol流域中部,大多数地下水资源赋存于前寒武基底的断裂含水层中。对地下水资源的特征描述很少,本研究第一次试图更好地描述含水层的机理和水文地质特性。综合采用水文地球化学和测压方法确定了浅层风化含水层和深层裂隙/断裂含水层的补给量变化和地下水成因。在流域尺度上,地下水流向尼日尔河的通量主要受地形的影响,没有地下水水位的长期趋势数据(1980–2009年)。水文地球化学印记主要受钙质重碳酸盐至镁(70%)类型支配。显示了从含有CO2及低矿化度水(花岗岩类岩石及蚀变重矿物)开放环境向含有矿化度较高的水(片岩)更封闭的环境演化过程。稳定同位素(δ18O, δ2H)分析表明,有两个主要的地下水补给机理:(1)几乎没有降雨后分馏印记的直接补给及(2)来自蒸发的地表水及/或河道河床的间接补给。地下水氚含量表明,补给主要是近代的,年龄少于50年(3H > 3 TU),只有10%的补给不是过去几十年的补给。通过氚各自的值估算出来的地下水更新率中间值相当于1.3% year−1,接近于20世纪80年代地下水样品确定的中间值,因此,表明没有重大的长期变化。

Caracterização dos mecanismos de recarga em um aquífero de embasamento Pré-cambriano no semiárido no Sudoeste de Niger

Resumo

Na parte central da bacia semiárida de Dargol no Sudoeste de Niger, a maior parte dos recursos hídricos subterrâneos está contido em um aquífero fraturado de embasamento Pré-cambriano. Os recursos hídricos subterrâneos estão mal caracterizados e esse estudo é a primeira tentativa de descrever melhor os mecanismos de reacarga e o comportamento hidrogeoquímico dos aquíferos. Métodos hidrogeoquímicos e piezométricos foram combinados para determinar mudanças na taxa de recarga e origem das águas subterrâneas para os aquíferos intemperizado raso e fraturado/fissurado profundo. Na escala da bacia, os fluxos de águas subterrâneas em direção ao Rio Niger são influenciados principalmente pela topografia, sem tendência de longo termo visível nos níveis de águas subterrâneas (1980–2009). A assinatura hidrogeoquímica é dominada pelo tipo bicarbonatada-cálcica para magnesiana (70%). Isso demonstra a evolução de um meio ambiente aberto com CO2 e água de baixa mineralização (granitoide, alteritos) em direção a um meio ambiente mais confinado com águas mais mineralizadas (xistos). Analise de isótopos estáveis (δ18O, δ2H) da água sugere dois principais mecanismos de recarga: (1) recarga direta sem assinatura pós-precipitação próxima e (2) recarga indireta de águas superficiais evaporadas e/ou leito do canal-córrego. O conteúdo de trítio indica que a recarga é mais recente, com idades menores que 50 anos (3H > 3 TU), com apenas 10% indicando baixa ou não recarga nas décadas passadas. O valor médio da taxa de renovação de águas subterrâneas estimado dos valores individuais de trítio é equivalente a 1.3% anos−1, próximo ao determinado para amostras de águas subterrâneas datando do começo da década de 80, assim indicando mudança de longo termo não mensurável.

Introduction

Since the 1970s, the Sahelian regions of West Africa have experienced climatic variability marked by recurrent droughts (Nicholson 2001; L’Hôte et al. 2002; Sagna et al. 2015). These drought periods are remarkable for their duration at the global scale, and characterized by a reduction of about 20% in precipitation (Taylor et al. 2002; Panthou et al. 2014). In the central Sahel, this has led to a degradation of soil surface conditions and lasting changes in the water balance (Leduc et al. 2001; Séguis et al. 2004; Mahé 2009). The high population growth (of about 2–3% year−1) and the estimated extension of cultivated areas (5% in 1960 to 14% in 1990) has been associated to a 28% loss of forest cover (Taylor et al. 2002) and have thus accentuated the effect of land clearance on soil crusting (Favreau et al. 2009). Hydrological studies have also shown the sustained indirect impact of these anthropogenic pressures on surface waters—for example, their effects are reflected in the increase in river flows despite the recurrence of droughts (Mahé et al. 2003). As a consequence of intense and persistent climate and environmental disturbances, it is necessary to consider how these changes affect groundwater (Lapworth et al. 2013, Ibrahim et al. 2014). In Niger, the impact of these changes has been assessed in unconfined sedimentary aquifers (Leduc et al. 2001, 2008). Higher piezometric levels observed since the mid-1960s are interpreted as the response to anthropogenic degradations of the vegetation cover, despite the decrease of rainfall (Favreau et al. 2009, 2012).

In the crystalline basement zone, the long-term dynamics of groundwater resources remain unknown. Filippi et al. (1990) showed that in the neighboring crystalline basement region of Burkina Faso, plurimetric groundwater-level fluctuations are caused by the seasonality of rainfall. For a large Sahelian catchment (20,000 km2) in Burkina-Faso, a clear link has been established (Mahé 2009) between the degradation of soil surface characteristics and an increase in Hortonian surface water flows. In the basement area of northeastern Mali, Gardelle et al. (2010) observed an increase in pond areas which could have an impact on the dynamics of underlying aquifers, if focused recharge occurs similarly to observations made in the sedimentary context of the neighboring Sahelian zones.

Previous research highlights the complexity of hydrogeological studies in semi-arid zones where the different terms of the hydrological balance are particularly sensitive to small environmental changes and where the response of the impacted systems can be counter-intuitive. Despite these difficulties, hydrogeological studies in crystalline basement zones are of primary importance. The uncertainty on water availability is generally more pronounced, since groundwater resources are often smaller in volume and discontinuous in space. In the south-west of Niger, about 40% of the wells drilled in villages have low yield (<0.5m3 h−1). Moreover, 55% of the productive wells are abandoned due to chemical problems, as contaminant concentrations (e.g. nitrate, arsenic, and fluoride) exceed WHO drinking-water standards (Ousmane et al. 2012). These difficulties of access to drinking water, coupled with the low mean productivity, explain why the region’s population is locally affected by severe water shortages during the long dry season. In the semi-arid crystalline basement regions, the main challenge lies in a better hydrogeological characterization in order to better understand the groundwater dynamics, not only in terms of recharge, but also in terms of groundwater quality.

Previous studies carried out in the crystalline area of south-west Niger revealed that recharge occurs following two mechanisms: (1) direct and diffuse recharge in some more permeable zones of the landscape (e.g. sand dunes) and, (2) indirect and punctual recharge in topographical depressions (Ousmane 1988; Girard et al. 1997; Abdou Babaye 2012) where water accumulates during the rainy season. Due to the spatial and temporal variability and the various processes associated with the recharge, estimating the rate of renewal of groundwater is usually challenging in semi-arid areas (De Vries and Simmers 2002). This complexity is accentuated by the fact that indirect localized recharge is often the dominant recharge mechanism, sometimes at low frequencies (Gaultier 2004; Bajjali 2008). Quantification of the localized recharge requires accurate and localized information, which, often over the long term, is unavailable. To compensate for the low density of observations, a range of techniques is commonly used to explain the link between the dynamics of the aquifers and climate and environmental changes (Ngounou Ngatcha et al. 2001, Stadler et al. 2010, Bouragba et al. 2011). In arid and semi-arid regions, multiple tracer approaches are preferred to estimate localized recharge, compared to the regional hydrological water budget method for which accurate quantification of this type of recharge remains difficult (De Vries and Simmers 2002; Scanlon et al. 2006). In complex water–rock-interaction media, isotopes of the water molecule (18O, 2H, 3H) have proven to be valuable tracers of the subsurface flow (among others: Aggarwal and Gat 2005; Clark 2015; Solder et al. 2016; Cook et al. 2017).

This study applies hydrodynamics data (groundwater levels, surface water levels) and hydrogeochemical data (major ions, water isotopes) that will be combined to specify, over the long term, the spatio-temporal dynamics of groundwater fluxes in the semi-arid southwest of Niger. This will achieve better management of the resource.

Case study

Geographic and climatic context

The study area is located in the south-western part of Niger, between longitudes 0°30′ and 0°50′ east and latitudes 13°45 ‘and 14°20’ north. With an area of 900 km2, the study area occupies the central part of the Dargol basin, which is a tributary on the right bank of the Niger River (Fig. 1). The climate is Sahelian with a rainy season extending from June through September. The extreme daily temperature values observed over the period 1999–2008 range from 15 °C (6 am) to 43 °C (6 pm), with an annual average of 29 °C. The relative humidity of the air shows an annual variation of 18% (March) to 90% (August). The mean annual rainfall at the Tera rainfall station is 409 mm (1999–2008), while the mean annual evapotranspiration calculated from the Penman method for the corresponding period is 2,000 mm (i.e. five times the annual rainfall). 25% of precipitation is from high-intensity events (Panthou et al. 2014) that produce rapid Hortonian flows (Amani and Nguetora 2002).
Fig. 1

Location and geological context of the study area (Adapted from Dupuis et al. 1991) within the Niger River Basin in West Africa (inset)

The density and orientation of the drainage network is influenced by the topography, the lithology and the tectonic structure of the substratum. Taking its source in Burkina Faso, about 320 m above sea level (m asl), the temporary river Dargol and its main tributary the Tilim flow on the right bank of the Niger river (elevation 198 m asl at the station of Kakassi), about 90 km northwest of Niamey (Fig. 1). Its flow is linked to rainfall that begins in June and July and ends 3 months later with an average flow of 160 × 106 m3/year (1964–1994). There is also a chain of temporary pools or artificial water reservoirs, of which only the Téra Dam contains water continuously (available water volume of 7.7 × 106 m3 at its creation in 1981).

The relief of the area is relatively flat with isolated hills and fixed sand dunes inherited from more arid periods of the Quaternary. These morphological assemblages are notched by usually dry valleys (koris) which drain water towards depressions (valleys, ponds) during the rainy season. Degraded and sparse vegetation (savanna and steppe) occupies the plateaux, whereas the bottom of the valleys and depressions give way to denser woody vegetation. The rate of local population growth was about 3.9% per year over the period 2001 to 2012, with more than 95% of the population living in rainfed agriculture or extensive livestock farming (INS 2012).

Geological and hydrogeological context

The bedrock of the study area is formed by a Precambrian basement composed of rocks of the metamorphic belt (green rocks including pyroxenites, amphibolites, epidotites, chloritoschists, metabasalts, metagabbros, greywacke rocks, tuffs, rhyolitic breccias, micaschists, clayey schists, quartzitoschists). These rocks show a NNE–SSW orientation. They are separated by Eburnean granitoid bodies (granites, granodiorites and diorites; Machens 1973; Soumaila and Konate 2005; Fig. 1). In some locations, these massifs contain Archean relics (pegmatites, leptynites) (Machens 1973; Dupuis et al. 1991). Shallow formations made of alterite (5–50 m thick), alluvium and colluvium overlie bedrock formations. At the regional scale, four main fracture orientations constitute the major structural features (Abdou et al. 1998; Affaton et al. 2000; Abdou Babaye 2012): N20°−N50°; N60°–N90°; N120°– N140° and N350°–N15°.

In crystalline and metamorphic schists environments, aquifers are developed in the shallow weathered formations (alterites and alluvial deposits) and in the fractures/cracks of the deep basement. These two superimposed aquifers can be considered as connected in some areas (Dewandel et al. 2006). The weathered aquifer is formed by semi-permeable materials with high storage capacity. It covers the upper part of the deep aquifer corresponding to the highly fractured zone due to decompression of the bedrock with high permeability. The lower part of the deep aquifer corresponds to the compact bedrock only affected by fractures of tectonic origin (faults). This deep fractured aquifer is generally confined, whereas, in some locations, it is able to drain the upper horizons. Most fractures therefore contribute to aquifer transmissivity but are considered to have a low storage capacity.

In the study area, the depth to the water table varies according to the topography, the type of aquifer and season. The shallow weathered layers are located in the valleys (alluvium) and the plateaux (alterites). The depths to the water table, as observed in traditional wells dug in alluvial formations, vary from 1 to 10 m. In modern wells, drilled in alterites, the water depth ranges between 15 and 30 m. Upper parts of the aquifer are very reactive to rainfall fluctuations (Ousmane 1988) and can dry up during the dry season leading to recurring problems of water shortage. Deep aquifer parts are, in principle, more promising in terms of productivity and sustainability because of the role played by fractures in the drainage, storage and circulation of groundwater. As a result, these deep aquifers have been the subject of several village water drilling programs to cover the water needs of the population in the crystalline area. Analysis of the data from 140 boreholes (Fig. 1) from these programs (1980–2009) allowed for synthesis of the hydrogeological parameters of the study area. The depths of these boreholes vary from 36 to 120 m with an average of 60 m and a standard deviation of 20 m. The best water production flows are generally obtained between 30 and 65 m; 76% of the water is produced in the first 50 m, constituting the upper part of the bedrock (Abdou Babaye 2012). Beyond this depth, transmissive zones are rare. In granitic rocks, the depth limit to these transmissive zones rarely exceeds 50 m, whereas it can reach 60 m (and even slightly deeper) in green rocks (schists) due to the nature and thickness of their weathering products. Lithological diversity and the various tectonic phases influence the hydrogeological properties of basement aquifers and, of course, their productivity. Analysis of the drilling data reveals that water yields are generally low, ranging from 0.5 to 5 m3 h–1, with the majority of boreholes being limited to values below 2.5 m3 h–1. In green rocks, higher water yield (mean of 2.7 m3 h–1 versus 1.9 m3 h–1 in granitoids) are observed. Flow rates greater than 10 m3/h were obtained in boreholes drilled in fractured networks of shear zones.

Hydraulic conductivity values estimated by short-time (4 h) pumping tests range from 2.4 × 10−7 m/s to 3 × 10−5 m/s with averages of 3 × 10−6 and 1.1 × 10−5 m/s respectively observed on granitoids and green rocks. Ousmane (1988) showed statistically that green rocks are more productive with a drilling success rate (flow rate > 0.5 m3 h–1) of 87% compared to about 60% for granitoid rocks.

Methodology

Groundwater level surveys

In the study area, measurements of static groundwater levels date from the drilling operations. Unfortunately, there is no regular and official monitoring of groundwater levels in rural and even urban waterworks. The absence of a piezometric network and the low density of measurement points did not allow drawing a reliable piezometric map. However, measurements (with a centimeter accuracy) of groundwater levels have been carried out in various topographic (basin, slope, plateau) and geological (schists, granites, weathered rocks) contexts in both high and low groundwater-level periods. Groundwater levels were measured, at the beginning of the day before the pumping activities, at 30 points in May 2009 and 40 points in October–November 2009 and in April–May 2010. In addition to these seasonal campaigns, monthly surveys were made from October 2009 through January 2012. Almost all the measurements were undertaken in wells screened in the bedrock fractured aquifer. Only six measurements were made in wells drilled in the superficial layers (alterites and alluvium). All the wells being in operation, the measurements were made very early in the morning (between 5 and 8 am) in order to obtain a measurement in conditions close to the static level. To this recent dataset, about 20 older measurements from the early 1980s have been added to evaluate the long-term dynamics of the groundwater levels.

Groundwater sampling

Forty-five water samples, taken for classical chemical analyses, were taken from shallow wells, boreholes and surface waters. These samples were taken during periods of high (October–December 2009) and low (April–May 2010) groundwater levels. The sampling points were chosen to be representative of the main different geological contexts of the fractured aquifers in the study area. Nevertheless, for the purpose of comparison, three analyses concerned the shallow aquifer, and a surface-water sample was also collected at Tera Dam.

In addition, most of the points sampled during the period of high groundwater level were also sampled for isotopes analyses. Thus, 22 waterworks (20 boreholes and 2 wells) were sampled for oxygen-18/deuterium analyses, and 17 sampled for tritium analyses. 5 of these points are corresponding to those already sampled in the 1980s (Ousmane 1988). In addition, a surface-water sample (Tera Dam) was taken and analyzed for its water isotope content.

During each sampling survey, physico-chemical parameters (pH, temperature, electrical conductivity) were measured in situ. Measurements and sampling were carried out after at least 30 min of pumping in order to obtain groundwater samples more representative of groundwater in the aquifer.

Chemical and isotope analyses

Chemical analyses (carried out in the HGE-ULiege laboratory) have ionic balance errors varying around ±3%. Analyses (Ca2+, Mg2+, Na+, K+, Cl, NO3, S042−) were carried out using a Capillary Ion Analysis (C.I.A.) technique. Silica was determined by a Flame Atomic Absorption technique, while carbonate (CO32−) and bicarbonate (HCO3) ions were calculated after determination of pH and the complete alkalimetric title.

Analyses of the oxygen-18/deuterium pair were carried out by the IDES Laboratory of the University of Paris-Sud (Orsay, France) and those related to tritium in the HYDROISOTOP laboratory in Schweitenkirchen, Germany. The results of the analyses are expressed in δ‰ vs V-SMOW. The analytical errors are respectively ±0.2 δ‰ and ± 2 δ‰ for oxygen-18 and deuterium. For tritium (expressed as a tritium unit, TU), the accuracy is ±0.5 TU with a detection limit of 0.9 TU.

Rainfall data were obtained from the International Atomic Energy Agency (IAEA) network (GNIP 2011), from Niamey/ORSTOM stations (1992–1999) and IRI/Abdou Moumouni University of Niamey in Niger, located at an almost identical (190 m vs 225 m asl) altitude (Ousmane 1988; Girard 1993; Taupin et al. 2002; Guéro 2003; Tremoy et al. 2012).

Results

Hydrodynamics

The monthly monitoring of the static groundwater piezometric levels shows seasonal fluctuations subsequent to the beginning of the rainy season for all (n = 7) of the wells surveyed, both in the shallow part of the aquifer in the alterites (shallow wells) and in the deep fractured aquifer (boreholes). The water table decreases from the end of October to early November, reaching the lower levels in May (Fig. 2). From the first weeks of July, a rapid (alterites) or progressive (deep fractured aquifer) rise of groundwater levels is observed. Maxima are reached in August–September or even in October, 2 or 4 months after the start of the rainy season. The amplitude of fluctuations of groundwater levels between periods of high and low groundwater levels varies from 0.29 to 13.1 m. The highest amplitudes are observed in the upper areas (plateau, sand dune).
Fig. 2

Fluctuations in the depth to the water table in the shallow aquifer (W1) and in the deep fractured aquifer (B1, B21; see Fig. 11 and Fig. 12). Daily rainfall measured at the Tera station

Due to the low amount of data, it is not possible to draw a map with hydraulic head contours; however, main flow directions can be derived from the available data (Fig. 3).The map of the main groundwater flow directions in the deep aquifer (Fig. 3), drawn for the high groundwater level periods (October–November 2009), shows a strong influence of the topography. The directions of the groundwater flows are thus quite similar to that of surface waters, and influenced by the directions of the major fractures, north–east, south–east and west–east. Some small piezometric domes are observed in the west, south-west, north and center corresponding probably to preferential infiltration from pools and artificial reservoirs (dam). The lowest groundwater levels are observed in the southeastern part of the basin in the vicinity of the Tilim plain and its confluence with the flood plain of the Dargol River. Comparison of the groundwater levels from the 1980s to 2009 shows a decrease in about 65% of the boreholes (13 out of 20 boreholes; Fig. 4), while on the other hand, a significant increase is observed in some boreholes, with no clear spatial pattern observed.
Fig. 3

Map of the main groundwater flow directions (Nov, 2009) in the deep aquifer part. Spatial distribution of the tritium content in groundwater

Fig. 4

Changes in groundwater levels between 1980 and 2009

Hydrochemistry

The results of the chemical analyses (synthesis in Table 1; full analyses in Table S1 of the electronic supplementary material ESM) reveal that the groundwater is slightly acidic to neutral and characterized by a variable mineralization (ranging from 266 to 1,184 μS/cm). Dam water, sampled in high water conditions (18 December 2009) shows a low mineralization (143 μS/cm). The temperature of the water varies from 29 to 34 °C in all the groundwater samples. The highest values could be overestimated due to surface measurements in periods of high atmospheric temperatures.
Table 1

Synthesis of the results of chemical and isotopic analyses for the high groundwater level period (The complete data are provided in Table S1 of the ESM)

Sample type (n)

Cond 25°

pH

Ca2+

Mg2+

Na+

K+

Cl

SO42−

NO3

CO32−

HCO3

δ18O

δ2H

d

3H (TU)

Green rocks (12)

Min

385

6.43

18.72

10.51

19.71

2.12

2.67

3.69

0.3

0.93

146.81

−2.32

−11.6

4.78

<0.9

Max

1123.3

7.3

128.28

48.66

76.97

7.2

23.43

294.95

231.1

10.54

390.18

−4.49

−31.1

8.24

4.7

Mean

658.47

6.86

54.7

26.06

47.1

4.32

9.26

57.27

51.25

4.91

278.71

−3.55

−21.9

6.52

3.51

SD

255.60

0.24

33.81

12.37

19.92

1.86

7.211

88.56

69.29

2.37

64.345

0.87

6.96

1.36

1.41

Granitoids (13)

Min

350

6.1

21.8

5.51

13.06

0.01

3.5

2.22

1.05

0.94

100.41

−4.13

−0.09

−0.39

1.4

Max

1184.5

7.1

108.1

26.86

84.06

5.17

45.22

40.67

405.8

7.41

353.6

0.9

−27.6

9.56

5.9

Mean

594.65

6.63

51.57

17.69

49.51

2.93

16.77

19.59

95.84

3.36

228.73

−2.94

−19.2

4.38

3.75

SD

298.98

0.27

26.02

15.39

21.90

1.517

13.3

12.25

124.6

1.93

80.43

1.52

8.24

4.64

1.21

Alterites (1)

719

7.1

74.78

27.38

33.62

6.26

42.10

21.75

138.24

3.95

211.25

0.19

−3.06

−4.58

Dam (1)

143.6

7.79

21.34

3.09

3.51

3.60

1.34

1.29

3.04

0.31

89.56

0.79

−1.99

−1.2

Chemical elements (mg/L), electrical conductivity (Cond) in μS/cm, deuterium excess (d), δ18O and δ2H (‰ V-SMOW). TU tritium unit, SD standard deviation

The deep and shallow aquifers show diversified facies (Fig. 5). The most represented groundwater facies is the calcium-magnesium bicarbonate facies (70% of the samples) followed by the calcium chloro-sulfate (25%) and sodium bicarbonate (5%). The chloro-sulfate calcium signature reflects the evolution of the anions of the bicarbonate pole towards the chloride-nitrate pole. This evolution is mainly due to an increase in nitrate, and secondarily by the addition of sulphates in the evolution towards a third facies. Chloride-nitrate water is interpreted as a characteristic of a recently recharged aquifer. On the other hand, the evolution of the waters of the bicarbonate calco-magnesian zone towards the bicarbonate sodium domain could be a sign of some aging of the waters due to the cation exchange indices between alkaline earths and alkalies (Diop and Tijani 2008).
Fig. 5

Chemical facies (Piper diagram) of the sampled waters (HWL: High water levels; LWL: Low water levels)

Isotopic signal of the local rainfall

The local meteoric waterline (LMWL), based on Niamey’s monthly rainfall (Taupin et al. 2003), has the following equation (1992–1999, n = 28 and R2 = 0.97):
$$ {\updelta}^2\mathrm{H}=7.7\ {\updelta}^{18}\mathrm{O}+5.4 $$
(1)
The slope of this LMWL (7.7) is lower than that of the World Meteorological Water Line (8). This characteristic is interpreted in the Sahel as the result of slightly evaporated precipitation during rainfall, in accordance with the interpretation of the monitoring of isotope concentrations of atmospheric water vapor in Niamey (Tremoy et al. 2012). The origin of precipitation is determined from the values of the deuterium excess:
$$ d={\updelta}^2\mathrm{H}-8\ {\updelta}^{18}\mathrm{O} $$
(2)

Tritium concentrations in precipitation show a series of peaks following the first nuclear tests in early 1950s. Current tritium concentrations are of 5–10 TU in the precipitation in the northern hemisphere (GNIP 2011). In southwestern Niger, the most recent measurements were ~7 TU in 1994 and ~4 TU in 2006 (Favreau et al. 2002; Favreau, IRD France, personal communication, 2016).

The weighted average of the measures, the rainy period number by month, collected in Niger and Burkina Faso between 1998 and 2008 is 5.2 TU (Ousmane 1988; Favreau et al. 2002; Guéro 2003; Yaméogo 2008). This value represents the isotopic signature of the current rains, and therefore the input function of the hydrological system. Given the half-life of tritium (12.41 years), groundwater infiltrated before the 1950s (average concentration estimated at 5 TU in precipitation) would have in 2009 (sampling date) a tritium concentration of 0.2 TU, therefore lower than the detection threshold.

Isotopes in groundwaters

Stable-isotope compositions of 22 groundwater samples, including two samples from wells dug in alterites, range from −4.5 to +0.9‰ for oxygen-18 and –31 to −0.1‰ for deuterium. The overall mean values and standard deviations are respectively −3.0 and 1.5‰ for oxygen-18 and −19 and 8.2‰ for deuterium. Considering only deep bottom aquifers (17 values), mean values of −3.6 and − 23‰ for oxygen-18 and deuterium respectively are found. δ18O levels are 36% higher than −3‰, while 77% have deuterium values lower than −25‰. These results highlight the influence of different lithologies and the different groundwater pathways in the area.

The calculated deuterium excess (d) values for groundwater are less than 10 (Table 1). This shows that the air masses that generated Niger’s precipitation and eventually groundwater recharge originate from the Guinean monsoon, with a marked effect of the re-evaporation of rainfall (Tremoy et al. 2012).

The isotopic composition of groundwater in arid regions can be considerably modified by evaporation in comparison to isotopic composition of the local precipitation (Clark and Fritz 1997; Favreau et al. 2002). However, despite high actual evapotranspiration (AET) values, it is possible to have groundwater with isotopic compositions close to the average precipitation composition, demonstrating rapid recharge flows to aquifers (Acheamponga and Hess 2000; Goni 2006). By plotting the sampled points in the diagram δ2H vs δ18O (Fig. 6), it is observed that all points are organized around the two following linear regression lines:
$$ {\updelta}^2\mathrm{H}=7.5\ {\updelta}^{18}\mathrm{O}+4.5\kern0.5em {R}^2=0.83\ \left(\mathrm{Samples}\ \mathrm{from}\ \mathrm{the}\ \mathrm{deep}\ \mathrm{aquifers},n=17\right) $$
(3)
$$ {\updelta}^22\mathrm{H}=4.8{\updelta}^{18}\mathrm{O}-4\kern0.5em {R}^2=0.95\kern0.75em \left(\mathrm{Samplesfromtheshallowaquifers},n=5\right) $$
(4)
Fig. 6

Graph of δ2H as a function of δ18O for groundwater and surface-water samples

The 17 points corresponding to the samples taken in the deep aquifers are aligned around line (3) in Fig. 6 (expressed by Eq. 3) with a slope of 7.5, very close to the slope of the line representative of the local precipitation (LMWL). This suggests rapid recharge to the aquifers without marked evaporation effects as shown in Figs. 7, 8, and 9 and confirmed by the computation of the water renewal rate for these aquifers. Three of these boreholes (circled points in Fig. 6) deviate from the LMWL and are below the evaporation line (4) in Fig. 6 (expressed by Eq. 4). This deviation from the LMWL could indicate enrichment by the evaporation process from surface water. The position of these boreholes in the interdunary basins and in sandy zones, which are favorable sites for the concentration of runoff water, the infiltration (Aranyossy and Gaye 1992; Ngounou Ngatcha et al. 2001) and the evaporation could explain their evaporated intermediate composition. Contrary to the observations made in the neighboring sedimentary zone (Favreau et al. 2002), the deeper part of the aquifers seem to be able to hold slightly evaporated waters. Desconnets et al. (1997) show that in the sedimentary zones of the Sahelian regions, ~80–90% of the water accumulated in the ponds infiltrate and contribute to the rapid recharge of the aquifers. On the other hand, in the basement area, clayey-sandy alterites have an effect on the infiltration rate due to their low permeability.
Fig. 7

Graph of magnesium (Mg2+) as a function of oxygen-18 (δ18O) in groundwater. This relationship makes it possible to distinguish deep and confined schist aquifers, from granitic aquifers overlaid by sandy clay weathering products more permeable than those found above the schist formations

Fig. 8

Bivariate plot showing the relationship between the saturation index of calcite (SIC) and dolomite (SID) for the groundwater samples

Fig. 9

Equilibrium diagram of albite-montmorillonite-kaolinite-gibbsite (at 25 °C) for groundwaters in granitoid, green rocks and alterites

Five points are aligned with the straight line (4) in Fig. 6, with a slope of 4.8, indicating an enrichment by evaporation of the surface water that supplies the aquifers by the indirect recharge processes (Abdalla 2009). These five points correspond to wells collecting shallow weathered layers and shallow boreholes screened at the interface between the fractured horizon and grainy weathered horizon. This evaporation line (4) intersects the LMWL through a point of coordinates −4.1 and − 23.1‰ corresponding to the mean weighted rainfall composition at the Niamey station. The fact that this point belongs to the evaporation line (2) in Fig. 6 indicates that the waters of the shallow aquifers come from the current rains which have undergone evaporation at the surface or in the first meters of the ground. In semi-arid zones, partial evaporation of water in the first meters of the soil is the basis of isotopic enrichment (Edmunds et al. 2002; Stadler et al. 2010).

Tritium activity in the bedrock aquifers shows a wide range of values. The concentrations ranged from <0.9 (detection limit) to 5.9 TU, with an average of 3.69 ± 1.22 TU. Generally speaking, 82% of the Liptako basement groundwater samples have tritium levels of ≥3 TU, of which 64% are >4 TU. This implies that bedrock aquifers contain an important part of the waters that were infiltrated recently, i.e. over the past 50 years. Nevertheless, there are two older water samples with low tritium content (<0.9 TU, 1.4 ± 0.6 TU). High nitrate concentration in this part of the aquifer may be inherited from soil nitrogen sources and subsequent leaching occurring during high recharge events (Schiewede et al. 2005; Favreau et al. 2009).

Waters with tritium levels close to those of recent precipitation (3–4.7 ± 0.7 TU) are found in both deep and shallow aquifers. These two levels are distinguished according to their oxygen-18 contents which are relatively enriched in the more shallow aquifers. These shallow aquifers can be characterized by an indirect and rapid mode of recharge through the major bed of koris or laterally by evaporated surface waters. Thirteen older groundwater samples, within the limits of the Dargol basin and dating back to the early 1980s (Ousmane 1988), were also considered for an estimate of long-term changes in the renewal rates based on tritium data.

The diagram of δ18O vs 3H highlights the presence of three groups of water belonging to different periods of recharge (Fig. 10). Thus, low-tritiated waters (<3 TU) belong to the deeper parts of aquifers characterized by higher oxygen-18 depletion. This older signal could correspond to a recharge during the wetter period of the mid-twentieth century (B12 < 0.9 TU). It may also be interpreted by a slow and diffuse mode of recharge probably induced by rainwater infiltration through the altered zones.
Fig. 10

Graph of 3H content as a function of δ18O in groundwater

Discussion

Hydrochemical evolution of the groundwater

The hydrogeochemical signature of waters can be used to trace water–aquifer matrix exchanges. An increase in groundwater bicarbonate (and in pH values) results from higher soil CO2 partial pressure supplied by the infiltration water, assuming there is a closed system in the deeper part of the aquifer. This trend is consistent with the results (see Fig. S1 of the ESM) indicating a positive correlation between HCO3 and (Na + K)-Cl in contrast to (Ca + Mg)-(HCO3 + SO4) which shows a weaker negative correlation. The latter could be due to the different origins of the Ca2+ and Mg2+ ions, because the ion exchange reaction supposes the replacement of these ions by Na+ as a function of aging of the water (see Fig. S2 of the ESM). This principle is clearly proved through the diagram (Na + K) - Cl vs (Ca + Mg) - (HCO3 + SO4) made from the points sampled in the dry season. The results of Principal Component (PC) analysis using the same water samples (Abdou Babaye 2012) show the existence of two main sources of ion production. The first subgroup (Ca2+, NO3, Cl) indicates the influence of anthropogenic activities (septic tanks, animal excrement around wells) in the mineralization of water, which is partly confirmed by the presence of NO3 (De La Vaissière 2006; Diaw 2008; Yaméogo 2008). It has been shown that the Mg2+, Na+, SO42− cluster revealed that these ions would come from a different process than pollution through the infiltration of recent waters (Abdou Babaye 2012). The Mg2+ ion in this group shows that some of the waters with a calco-magnesian bicarbonate facies may be associated to old waters (Fig. 10).

The relationship between Mg2+ and δ18O clearly discriminates waters of the superficial aquifers from those of the deeper aquifers (Fig. 7). This relationship makes it possible to distinguish deep and confined schist aquifers, from granitic aquifers overlaid by sandy clay weathering products more permeable than those found above the schist formations. A clear increase in magnesium content with depth can be observed, except for the boreholes drilled in the granite basement. This increase, and the oxygen-18 among the most depleted in the groundwater samples, would then indicate the relationship between the geochemical process of hydrolysis and the more or less intense water–rock interactions.

The saturation index (SI) of calcite and dolomite (Fig. 8) and the equilibrium diagram (Fig. 9) show that the hydrodynamic functioning of these systems is influenced by the lithology of the reservoirs (granitoids, schists, alterites). Thus, recent waters (group 1 in Fig. 8), with a high value of pCO2 and very negative values of the saturation index (sub-saturation), are found in the wells and boreholes close to the Dargol flood plain. In contrast, waters with a longer residence time (group 3 in Fig. 8) are found on the plateau, the dunes or in the schist formations overlaid by a thick weathered layer. For the latter, one can deduce a slower circulation rate of the waters with relatively depleted contents of heavy isotopes (δ18O < −4‰) and tritium (<3 TU). The isotopic signature of these waters proves that the diffuse recharge is low and that the water resulting from the indirect recharge (from topographic depression) would takes time (≥ 50 years based on the lower tritium values) before reaching this part of the aquifer. Regarding the geochemical water–rock interaction (Fig. 9), waters are in the stability domains of kaolinite and montmorillonite. In the equilibrium diagram, this distribution of points is clearly influenced by the lithological nature, or by the distance to the recharge zone. The samples show an evolution from an open environment (groundwater in chemical equilibrium with kaolinite), where groundwaters are less mineralized (granitoid, alterites), towards a confined environment (groundwater more in equilibrium with montmorillonite) as in green rocks with naturally more mineralized waters. Groundwaters are becoming ‘older’ when passing from the kaolinite equilibrium towards the montmorillonite equilibrium showing confined conditions and probably associated low permeability values. This confinement is mainly related to the thick clayey-sandy alterite layers found in the schists zones, but also to the low-permeabability filling of some fractures.

Recharge rates and changes in groundwater storage

The annual and interannual fluctuations in groundwater levels and the hydrogeochemical behavior of the aquifers reveal the complexity of recharge processes in this semi-arid zone. The litho-tectonic context of the study area accentuates spatial heterogeneity in the hydraulic properties of aquifers due to the variable nature of the weathering products and the degree of fracturation of the surrounding rocks. Variations in groundwater levels are greater in boreholes located at high elevations (plateaux, dunes, rather remote from koris) than in those located in the immediate vicinity of floodplains of koris or ponds (Ousmane 1988). In schists or dune areas, the characteristics of the land surface results in low diffuse infiltration. The large thickness of the unsaturated zone, the clogging of the ponds and the low infiltration favor the loss of water by evaporation (Fig. 11).
Fig. 11

Geological cross-section (CD axis in Fig. 3) indicating the confinement of the deep aquifer under the alterite layer. Corresponding isotopic contents and electrical conductivity (EC) of groundwater measured in boreholes are reported

These arguments corroborate those suggested by the hydrogeochemical study discussed above. Heavy isotope depletion (δ18O < −4‰) and lower tritium content (<3 TU) suggest that the aquifers contain a significant component of older water and that recharge by direct infiltration of rainfall is low. The hypotheses on the rate of water renewal are verified by using a simple analytical model applied in the semi-arid zone (Leduc et al. 2000; Le Gal La Salle et al. 2001; Favreau et al. 2002) where the water renewal rate is high. This model makes it possible to calculate not only renewal rates in 2009 but also possible changes since the early 1980s (Table 2). Thus, the following equation can be used to calculate the average tritium content in an aquifer in a year i:
$$ {\mathrm{Av}}_i=\left(1-{\mathrm{Tr}}_i\right){\mathrm{Av}}_{i-1}{e}^{-\left(\frac{\ln 2}{T}\right)}+{\mathrm{Tr}}_i{\mathrm{Ap}}_i $$
(5)
where Tr i is the renewal rate of the water in the aquifer, T the half-live of tritium (12.43 years) and Ap i the tritium content in precipitation for the year i.
Table 2

Comparison of tritium content in boreholes in 1980–1981 (in Ousmane 1988) and 2009 (this study)

Locality

Ousmane (1988)

This study

Renewal rate in 1980 (% per year)

Renewal rate in 2009 (% per year)

Tritium (TU)

Date

Tritium (TU)

Date

Sirfikouara (B19)

8.1 ± 2.4

23/10/1980

4.1 ± 0.8

01/11/2009

0.44

1.25

Tera camp (B20)

15

17/06/1981

3.6 ± 1.1

31/10/2009

0.86

1.0

Tera pont (B21)

15

16/06/1981

3 ± 0.5

31/10/2009

0.86

0.78

Toumbindé (B7)

15

24/10/1980

4.7 ± 0.7

27/10/2009

0.86

1.60

Yanga (B12)

41 ± 7

24/10/1980

< 0.9

25/10/2009

3.3

<0.2

Dam

15

19/03/1981

5.2 ± 0.8

18/12/2009

Rainfall

15a

31/08/1980

5.21b

1998-2008

aNiamey-Université Station

bMean tritium concentration in rainfall (period 1998–2008; Dosso and Niamey stations) and from Burkina Faso (Ouagadougou Station)

Given the uncertainties associated with the recharge and tritium concentration in the recharge, and the possible variations in pumping or volume of the aquifer, the only significant variations between 1980 and 1981 and the end of 2009 are observed for the wells Sirfikouara (B19) and Toumbindé B7) with respectively 2 and 3 times the renewal rate. The aquifer remained stable at wells B20 and B21. The proximity of these wells (B19, B20, B21) to the riverbed of the koris and especially the permeability of the shallow weathered materials (sands and granitic sand) facilitate the infiltration of water on the one hand, but also the direct hydraulic relationship between the superficial and deep aquifers on the other hand. This is highlighted on the cross-section AB (Figs. 3 and 12) showing the hydraulic equilibrium between these two superimposed aquifers, contrary to the phenomenon observed in schists formations (cf. Fig. 11).
Fig. 12

Geological cross-section (AB axis in Fig. 3) showing the direct hydraulic connection between both the shallow and deep aquifers in the granitic zone. Corresponding isotopic contents and EC of groundwater measured in boreholes are reported

In the Yanga area (B12; Fig. 11), an apparent decrease in the renewal rate is observed. The exceptional characteristics of waters sampled in B12 are mainly their high mineralization (>1,100 μS/cm) but also their tritium content below the detection limit (<0.9 TU). These samples also belong to the family of waters close to saturation (cf. Fig. 8) corresponding to confined aquifers, with slow transit time leading to a low renewal rate or even no renewal during the years of drought. All these elements suggest that these waters have infiltrated very rapidly in a more humid period and at a lower temperature than that of the current climate. This is consistent with the theory of mixing of water of ancient origin prior to 1954 (Girard et al. 1997), as suggested by Ousmane (1988) from the activities of 14C.

The results of the model do not show a clear trend in the evolution of the aquifer renewal rate or in the storage in the aquifer (see Fig. 3). Comparison of renewal rates computed on the long-term is appropriate for groundwaters (McDonnell 2017); the median value of individual renewal rates of the 1980-1981 data set of groundwater tritium is of 20% per decade, an estimate that can be considered close, considering the high interannual variability in recharge in the Sahel (Favreau et al. 2009), to the one inferred (13% per decade) from the groundwater sampling and analyses performed in 2009, approximately three decades later.

Recharge conceptual model

Understanding the different recharge processes is an essential step for further assessment of the hydrodynamic functioning of the aquifers. Studies on the recharge mechanisms in arid and semi-arid environments in Africa have shown the importance of the drainage network in this process (Desconnets et al. 1997; Leduc et al. 1996). Temporary watercourses and endorheic pools can constitute preferential recharge zones. In basement areas, the concordance between the hydrographic network and the main tectonic fractures could enable a preferential infiltration towards the deep basement (Rana 1998).

The spatial distribution of tritium contents (Fig. 3) shows that the groundwater with high tritium concentrations is located near the koris beds. The tritium content in groundwater also decreases as the distance of the sampling points to the topographic depressions zones increases. Two assumptions could be invoked for explaining the presence of older waters in the upper areas. One could argue that the recharge occurs by direct diffuse percolation of the rain through the variably saturated zone. Its relative low permeability would explain long travel times. This is however unlikely the case considering the large thickness over the saturated zone and the dry conditions favoring evaporation of the infiltrated waters (Favreau et al. 2002). Moreover, some observations are in contradiction with this hypothesis showing a current isotopic signature despite the thickness and low permeability of the overlying cover (see Fig. 8). In the basement area, groundwater flow is generally influenced by the density of fractures, but especially by the deep fractures which constitute the preferential pathways for groundwater. This leads to the second hypothesis, which states that major fractures are the preferential pathways for the rapid drainage of recharge waters. Thus, recent waters are found in boreholes screened in major fractures, while old waters are located in less permeable areas characterized by secondary fractures (Diop and Tijani 2008). Ousmane (1988) showed that the isotopic contents of the Sahelian basement water generally show a significant heterogeneity since each isolated fracture constitutes a distinct aquifer system recharged locally by vertical infiltration. In the Kobio basin, located about 100 km south-east of Tera (see Fig. 1), Girard et al. (1997) have highlighted the importance of the fracture network in the recharge process, but also the continuity of the aquifer due to the fracture network density.

The interconnection of fractures and the permeability of filling materials locally determine the continuity of the aquifer (Ball et al. 2010). However, the boreholes may belong to different hydraulic systems, isolated by fractures filled by clays acting as a low-permeability barrier (Ousmane et al. 1983). The low tritium (below the detection limit) and low heavy isotope (δ18O of −7.4‰) content are related to the litho-structural context and to the recharge mechanism of the upper zones (Girard et al. 1997). These were demonstrated by the results of the present study through the isotopic heterogeneity accentuated by the hydraulic discontinuity due to the faults which do not favor the continuity of circulation of water beyond this zone (Ousmane 1988; Nkotagu 1996). The boreholes (B1, B11) located along the NE–SW major fault where the Tilim River flows show an isotopic homogeneity (18O and 3H), whereas those located on either side of this fault show different isotopic signatures (see Fig. 8, Fig. 3). These very close boreholes capture different fractures, fractures connected to the major fault with recent waters due to rapid and recharge, and secondary fractures with isolated old waters (Diop and Tijani 2008). The non-connection of the latter with the major fractures and the drainage network give these waters specific isotopic signatures; thus, these fractures can only receive water from vertical infiltration with transit times up to 30–60 years under fallow soils in arid regions (Ibrahim et al. 2014). According to Lapworth et al. (2013), the transit time through the unsaturated zone can reach 100 years in semi-arid regions. This delay may, however be shortened by the low thickness of the unsaturated zone or the high conductivity of fractures (Stadler et al. 2010) leading to a very short transit time.

It should be noted that in basement areas of humid regions, stable-isotope compositions are relatively homogeneous in fractured aquifers due to the high precipitation and water stock in the unsaturated zone (Adiaffi et al. 2009; Lapworth et al. 2013). This zone acts as a buffer to homogenize by mixing the successive infiltration waters and minimize the chemical heterogeneity in the saturated zone (Lapworth et al. 2013). The spatial heterogeneity of the isotopic composition of groundwater in arid and semi-arid regions (Leduc et al. 1996; Le Gal La Salle et al. 2001) is due to the spatial and temporal variations of the recharge. In addition, rainfall events from the core of the rainy season (mid-July–mid-September) are known to usually result in more depleted values in their stable isotopic composition (Ousmane 1988; Goni et al. 2001; Taupin et al. 2002, 2003; Favreau et al. 2002; Saravana et al. 2009; Massing and Tang 2010). Processes favoring infiltration during those periods could influence the groundwater isotopic composition.

Impact of vegetation cover and climate variability

Several studies carried out in arid zones have demonstrated the influence of climatic variability and changes in vegetation cover on water resources (Leduc et al. 2001; Favreau et al. 2009; Ibrahim et al. 2014). In Niger, degradation of vegetation cover and increase in growing areas have resulted in an increase in runoff, despite a decline of about 20% in annual rainfall (Taylor et al. 2002). Thus, the global runoff coefficient (i.e. ratio between total rainfall and measured total runoff at a discharge point of the considered basin) calculated on the Dargol at Tera has increased from 6.2% (before 1972) to 8.4% over the period from 1972 to 1992. Soil crusting and closure of macropores partly explain the predominance of localized recharge, leading to an increase in groundwater levels in the immediate vicinity of watercourses in the Continental Terminal (CT) aquifer (Favreau et al. 2002), ~150 km east of the study area. This process could explain the large increase in the groundwater level observed in the boreholes (W1 and B9; Fig. 4) localized in topographic depressions. Contrary to the observation made in the sedimentary areas, basement aquifers generally show a piezometric decline since the early 1980s (Fig. 4). For the Burkina-Faso basement area, Filippi et al. (1990) showed that the decrease in rainfall results in a decrease in groundwater levels. The sensitivity of the aquifers to rainfall fluctuations is due to the rapid mode of recharge through open fractures on the one hand and to the low capacity of soil aquifers to store a lot of water on the other hand. These distributions distinguish them from regional sedimentary aquifers, where renewal rates are usually much lower (< 0.1% year−1 in the CT aquifer in SW Niger; Favreau et al. 2002) that can more easily mitigate the effect of climate change.

Conclusion

The combined interpretation of groundwater level and hydrogeochemical data provided new insights into the hydrodynamic functioning and recharge processes of the fissured and weathered aquifers of southwestern Niger. The shallow aquifers of alterites are unconfined, whereas those of the fissured and/or fractured basement are often confined. The degree of confinement of these layers depends on the covering materials that are genetically related by weathering to the type of bedrock. Thus, the permeability of the alteration products of the granitic rocks allows a good interaction (and therefore without confinement) between the superficial and deep aquifers, while on the other hand, the less permeable clayey alteration above the schist aquifers creates confined conditions in these aquifers. The chemical and isotopic data have provided useful information to trace the path followed by infiltration water and their residence time in aquifers. Thus, the origin and the recharge processes, and the transit time of the waters have been specified. The main conclusions drawn from this study can be summarized as follows:
  • The distribution of isotopic contents in groundwater shows spatial variations accentuated by the hydraulic discontinuity due to fractures. These variations confirm a relative independence of each fracture in the groundwater circulation.

  • The recharge of the aquifers follows two distinct mechanisms: a direct recharge from the precipitation that does not undergo significant evaporation for the majority of the boreholes, and an indirect recharge from highly evaporated surface water found in the wells getting water from the alterite or shallow aquifers;

  • Tritium contents confirm an important component of recent waters, which, given the depth of the screen depths (20–120 m), implies rapid infiltration through fractures or major faults superimposed on the surface drainage network. In the groundwater sector where tritium concentrations are <3 TU, the renewal rate may be lower. The evolution of radioactive decay in the waters of the boreholes sampled at the beginning of the 1980s provides further details on the role of diffuse infiltration in the recharge of the waters but also on the mixing process and changes in the renewal rate, and recharge on a pluri-decennial scale.

  • The water–rock interaction, through the saturation index and the equilibrium diagrams of the minerals, allowed to estimate the approximate age of the water and to distinguish the parts of the aquifers where the recharge is weak.

This study proves that the aquifer system is well connected to the land surface through the fracture system. In addition to the low hydraulic conductivity and storativity of these aquifers, this means that the aquifers are very sensitive both to drought and pollution. A long drought resulting from climate change could result in dramatic decline in water level and well drying. The potential impact of agricultural and industrial development on groundwater quality is also immediate and huge. As a result, protection against contamination remains a major challenge. Taking into account these two problems is thus of primary importance for the management of these groundwater resources.

Notes

Acknowledgements

This study was funded by the Belgian Technical Cooperation (BTC), the University of Liège (ULiege) and by The International Foundation for Science (IFS grant, 2008). Some technical support was also provided by the AMMA-Catch team at IRD in Niger. This would not have happened without the logistical help of Abdou Moumouni University (UAM) and we express our deep gratitude to them.

Supplementary material

10040_2018_1799_MOESM1_ESM.pdf (242 kb)
ESM 1 (PDF 242 kb)

References

  1. Abdalla OAE (2009) Groundwater recharge/discharge in semi-arid regions interpreted from isotope and chloride concentrations in North White Nile rift, Sudan. Hydrogeol J 17:679–692.  https://doi.org/10.1007/s10040-008-0388-9
  2. Abdou Babaye MS (2012) Evaluation des ressources en eau souterraine dans le bassin de Dargol (Liptako–Niger) [Groundwater resources estimates in the River Dargol basin (Liptako, Niger)]. PhD Thesis Univ. Liège, Belgium, 235 ppGoogle Scholar
  3. Abdou A, Bonnot H, Bory K, Chalamet D, St Martin M, Younfa I (1998) Notice explicative des cartesgéologiques du Liptako à 1/100 000 et 1/200 000 [Explanatory note of the geological maps of the Liptako region at scales of 1/100,000 and 1/200,000]. Ministry of Mines and Energy, Niamey, Niger, 64 ppGoogle Scholar
  4. Acheamponga SY, Hess JW (2000) Origin of the shallow groundwater system in the southern Voltaian Sedimentary Basin of Ghana: an isotopic approach. J Hydrol 233:37–53CrossRefGoogle Scholar
  5. Adiaffi A, Marlin C, Oga YMS, Massault M, Noret A, Biemi J (2009) Palaeoclimatic and deforestation effect on the coastal fresh groundwater resources of SE Ivory Coast from isotopic and chemical evidence. J Hydrol 369:130–141CrossRefGoogle Scholar
  6. Affaton P, Gaviglio P, Pharisat A (2000) Réactivation du craton Ouest-africain au Panafricain : paléocontraintes déduites de la fracturation des grès néoprotérozoïques de Karey Gorou (Niger, Afrique de l’Ouest) [Reactivation of the West-African craton during the Panafrican tectonic cycle: evidence of paleostresses recorded by brittle deformation in the Neoproterozoic sandstones of KareyGorou (Niger, West Africa)]. C R Sci Acad Paris 331:609–614Google Scholar
  7. Aggarwal, P. K., Gat J. R. and K. F. O Froehlich (eds) (2005) Isotopes in the water cycle: past, present and future of a developing science. Springer, Dordrecht, The NetherlandsGoogle Scholar
  8. Amani A, Nguetora M (2002) Evidence d’une modification du régime hydrologique du fleuve Niger à Niamey [Evidence of a change in the hydrological regime of the river Niger in Niamey]. IAHS Publ 274, IAHS, Wallingford, UK, pp 449–456Google Scholar
  9. Aranyossy JF, Gaye CB (1992) Use of the tritium thermonuclear peak in the deep unsaturated zone for quantitative estimate of aquifer recharge under semiarid conditions: first application in the Sahel. CR Acad Sci Paris 315(Ser II):637–643Google Scholar
  10. Bajjali W (2008) Evaluation of groundwater in a three-aquifer system in Ramtha area, Jordan: recharge mechanisms, hydraulic relationship and geochemical evolution. Hydrogeol J 16:1193–1205CrossRefGoogle Scholar
  11. Ball LB, GeS, Caine JS, Revil A, Jardani A (2010) Constraining fault-zone hydrogeology through integrated hydrological and geoelectrical analysis. Hydrogeol J 5:1057–1067Google Scholar
  12. Bouragba L, Mudry J, Bouchaou L, Hsissou Y, Krimissa M, Tagma T, Michelot JL (2011) Isotopes and groundwater management strategies under semi-arid area: case of the Souss upstream basin (Morocco). Appl Radiat Isot 69(2011):1084–1093CrossRefGoogle Scholar
  13. Clark ID (2015) Groundwater geochemistry and isotopes. CRC Press, Boca Raton, FLGoogle Scholar
  14. Clark I, Fritz P (1997) Environmental isotopes in hydrogeology. CRC Press, Boca Raton, 328 pGoogle Scholar
  15. Cook P, Dogramaci S, McCallum J, Hedley J (2017) Groundwater age, mixing and flow rates in the vicinity of large open pit mines, Pilbara region, northwestern Australia. Hydrogeol J 25:39–53CrossRefGoogle Scholar
  16. De La Vaissière R (2006) Etude de l’aquifère néogène du Bas-Dauphiné: apports de la géochimie et des isotopes dans le fonctionnement hydrogéologique du bassin de Valence (Drôme, Sud-Est de la France) [Study of the Neogene aquifer of the lower Dauphiné region: contribution of geochemistry and isotopes approaches in the hydrogeological functioning of the Valence basin (Drôme, southeastern France)]. PhD Thesis, University of Avignon and the Vaucluse region, France, 339 ppGoogle Scholar
  17. De Vries JJ, Simmers I (2002) Groundwater recharge: an overview of processes and challenges. Hydrogeol J 1:5–17CrossRefGoogle Scholar
  18. Desconnets JC, Taupin JD, Lebel T, Leduc C (1997) Hydrology of the Hapex-Sahel central super-site: surface water drainage and aquifer recharge through the pool systems. J Hydrol 188:155–178CrossRefGoogle Scholar
  19. Dewandel B, Lachassagne P, Wyns R, Maréchal JC, Krishnamurthy NS (2006) A generalized 3-D geological and hydrogeological conceptual model of granite aquifers controlled by single or multiphase weathering. J Hydrol 330:260–284CrossRefGoogle Scholar
  20. Diaw M (2008) Approche hydrochimique et isotopique de la relation eau de surface/nappe et du mode de recharge de la nappe alluviale dans l’estuaire et la basse vallée du fleuve Sénégal: identification des zones inondées par télédétection et par traçage isotopique [Hydrodynamics and isotopic approaches of the relationships between surface and groundwaters and of the recharge processes of the alluvial aquifer in the downstream part of the Senegal River: identifying flooded areas by remote sensing and isotopic tracers]. PhD Thesis, Univ. C.A. Dakar, Senegal, 178 ppGoogle Scholar
  21. Diop S, Tijani MN (2008) Assessing the basement aquifers of eastern Senegal. Hydrogeol J 16:1349–1369.  https://doi.org/10.1007/s10040-008-0353-7 CrossRefGoogle Scholar
  22. Dupuis D, Pons J, Prost AE (1991) Mise en place de plutons et caractérisation de la déformation birimienne au Niger occidental [Plutonic emplacement and Birimian deformation in western Niger]. CR Acad Sci Paris 312(II):769–776Google Scholar
  23. Edmunds WM, Fellman E, Goni IB, Prudhomme C (2002) Spatial and temporal distribution of groundwater recharge in northern Nigéria. Hydrogeol J 10:205–215CrossRefGoogle Scholar
  24. Favreau G, Leduc C, Marlin C, Guero A (2002) A rising piezometric depression in the Sahel (southwestern Niger). C R Geosci 334:395–401CrossRefGoogle Scholar
  25. Favreau G, Cappelaere B, Massuel S, Leblanc M, Boucher M, Boulain N, Leduc C (2009) Land clearing, climate variability, and water resources increase in semiarid Southwest Niger: a review. Water Resour Res 45:W00A16.  https://doi.org/10.1029/2007WR006785 CrossRefGoogle Scholar
  26. Favreau G, Nazoumou Y, Leblanc M, Guéro A, Goni IB (2012) Groundwater ressources increase in the Iullemmeden Basin, chap 7. In: Climate change effects on groundwater resources:a global synthesis of findings and recommendations, vol 27. Taylor and Francis, London, pp 113–128Google Scholar
  27. Filippi C, Milville F, Thiery D (1990) Evaluation of natural recharge to aquifers in the Sudan-Sahel climate using global hydrological modelling: application to ten sites in Burkina Faso. Hydrol Sci J 35:29–48CrossRefGoogle Scholar
  28. Gardelle J, Hiernaux P, Kergoat L, Grippa M (2010) Less rain, more water in ponds: a remote sensing study of the dynamics of surface waters from 1950 to present in pastoral Sahel (Gourma region, Mali). Hydrol Earth Syst Sci 14:309–324CrossRefGoogle Scholar
  29. Gaultier G (2004) Recharge et paléorecharge d’une nappe libreen milieu sahélien (Niger Oriental): approches géochimique et hydrodynamique [Recharge and paleorecharge of a superficial aquifer in the Sahel (southeast Niger): geochemistry and hydrodynamic approaches]. PhD Thesis, Univ. Paris Sud, France, 179 ppGoogle Scholar
  30. Girard P (1993) Techniques isotopiques (15N, 18O) appliquées à l’étude des nappes des altérites et du socle fracturé de l’ouest africain: étude de cas—l’Ouest du Niger [Isotopic techniques (15N, 18O) applied for groundwater in alterites and hard rock aquifers of West Africa: case study—West Niger]. PhD Thesis, Univ. Québec à Montréal, Canada, 141 ppGoogle Scholar
  31. Girard P, Hillaire-Marcel C, Oga MS (1997) Determining the recharge mode of Sahelian aquifers using water isotopes. J Hydrol 197:189–202CrossRefGoogle Scholar
  32. GNIP (2011) Global network of isotopes in precipitation. http://www-naweb.iaea.org/napc/ih/IHS_resources_isohis.html. Accessed January 2018
  33. Goni IB (2006) Tracing stable isotope values from meteoric water to groundwater in the southwestern part of the Chad basin. Hydrogeol J 14:742–752CrossRefGoogle Scholar
  34. Goni IB, Fellman, Edmunds WM (2001) Rainfall geochemistry in the Sahel region of northern Nigeria. Atmos Environ 35:4331–4339CrossRefGoogle Scholar
  35. Guéro A (2003) Étude des relations hydrauliques entre les différentes nappes du complexesédimentaire de la bordure sud-ouest du bassin des Iullemmeden (Niger): approches géochimique et hydrodynamique [Study of hydraulic relationships between the different aquifers of the sedimentary complex of the southwestern border of the Iullemmeden basin (Niger): hydrodynamics and geochemical approaches]. PhD Thesis, Univ. Paris XI Orsay, France, 265 ppGoogle Scholar
  36. Ibrahim M, Favreau G, Scanlon BR, Seidel IC, Le Coz M, Demarty J, Cappelaere B (2014) Long-term increase in diffuse groundwater recharge following expansion of rainfed cultivation in the Sahel, West Africa. Hydrogeol J.  https://doi.org/10.1007/s10040-014-1143-z
  37. INS (2012) Présentation des résultats préliminaires du quatrième recensement général de la population et de l’habitat (RGP/H) [Presentation of the preliminary results of the 4th general census of the population and living conditions (RGP/H)]. INS, Niamey, Niger, 9 ppGoogle Scholar
  38. L’Hôte Y, Mahé G, Some B, Triboulet JP (2002) Analysis of a Sahelian annual rainfall index from 1896 to 2000: the drought continues. Hydrol Sci 47:56–572Google Scholar
  39. Lapworth DJ, MacDonald AM, Tijani MN, Darling WG, Gooddy DC, Bonsor HC, Araguás-Araguás LJ (2013) Residence times of shallow groundwater in West Africa: implications for hydrogeology and resilience to future changes in climate. Hydrogeol J 21(3):673–686.  https://doi.org/10.1007/s10040-012-0925-4
  40. Le Gal La Salle C, Marlin C, Leduc C, Massault M, Favreau G (2001) Renewal rate estimation of groundwater based on radioactive tracers (3H et 14C) in an unconfined aquifer an a semi-arid area, Iullemede Basin. Niger J Hydrol 254:145–156CrossRefGoogle Scholar
  41. Leblanc M, Favreau G, Massuel S, Tweed S, Loireau M, Cappelaere B (2008) Land clearance and hydrological change in the Sahel: SW Niger. Glob Planet Chang 61(4):135–150Google Scholar
  42. Leduc C, Taupin J-D, Le Gal La Salle C (1996) Estimation de la recharge de la nappe phréatique du Continental Terminal (Niamey, Niger) à partir des teneurs en tritium [Recharge of the Continental Terminal water-table (Niamey, Niger) estimated from tritium measurements and modelling]. C R Acad Sci Paris 323(Série II a):599–605Google Scholar
  43. Leduc C, Sabljak S, Taupin J-D, Marlin C, Favreau G (2000) Estimation de la recharge de la nappe quaternaire dans le Nord-Ouest du bassin du lac Tchad (Niger oriental) à partir de mesures isotopiques [Recharge of the Quaternary water table in the northwestern Lake Chad basin (southeastern Niger) estimated from isotopes]. CR Acad Sci Paris 330:355–361CrossRefGoogle Scholar
  44. Leduc C, Favreau G, Schroeter P (2001) Long-term rise in a Sahelian water-table: the continental terminal in South-West Niger. J Hydrol 243:43–54CrossRefGoogle Scholar
  45. Machens E (1973) Contribution à l’étude des formations du socle cristallin et de la couverture sédiementaire de l’Ouest de la République du Niger. Mémo [Contribution to the study of the formations of the crystalline basement and the sedimentary cover of the West of the Republic of Niger. Memo] No. 82, BRGM, Orléans, France, 167 ppGoogle Scholar
  46. Mahé G (2009) Surface/groundwater interactions in the Bani and Nakambe rivers, tributaries of the Niger and Volta basins, West Africa. Hydrol Sci 54(4):704–712CrossRefGoogle Scholar
  47. Mahé G, Leduc C, Amani A, Paturel JE, Girard S, Servat E, Dezetter A (2003) Augmentation récente du ruissellement de surface en région soudano-sahélienne et impact sur les ressources en eau [Recent increase in runoff in the Sudano-Sahelian region and its impact on water resources]. In: Hydrology of Mediterranean and semiarid regions. IAHS Publ. 278, IAHS, Wallingford, UK, pp 215–222Google Scholar
  48. Massing O, Tang Z (2010) Recharge of the Quaternary aquifer of Lake Chad Basin estimated from Oxygen-18 (18O) and tritium (3H) isotopes. Am Sci 6(9):283–292Google Scholar
  49. McDonnell J (2017) Beyond the water balance. Nat Geosci 10. www.nature.com/naturegeoscience. Accessed December 2017
  50. Ngounou Ngatcha B, Mudry J, Wakponou A, Ekodeck GE, Njitchoua R, Sarrot-Reynaul J (2001) Le cordon sableux Limani-Yagoua, extrême-Nord Cameroun, et son rôlehydraulique [The Limani-Yagoua sandy belt, far North Cameroon, and its hydraulic role]. J Afr Earth Sci 32(4):889–898CrossRefGoogle Scholar
  51. Nicholson SE (2001) Climatic and environmental change in Africa during the last two centuries. Clim Res 17:123–144CrossRefGoogle Scholar
  52. Nkotagu (1996) Origins of high nitrate in groundwater in Tanzania. J Afr Earth Sci 21(4):471–478CrossRefGoogle Scholar
  53. Ousmane B (1988) Étude géochimique et isotopique des aquifères du socle de la bande sahélienne du Niger (Liptako, Sud-Maradi, Zinder-Est) [Geochemical and isotopic study of the basement aquifers of the Sahelian belt of Niger (Liptako, South Maradi, East Zinder]. PhD Thesis, Univ. Niamey, Niger, 175 ppGoogle Scholar
  54. Ousmane B, Fontes JC, Aranyossy JF, Joseph A (1983) Hydrologie isotopique et hydrochimie des aquifères discontinus de la bande sahélienne et de l’Aïr (Niger) [Isotope hydrology of the discontinuous aquifers of the Sahelian belt and the Aïr (Niger)]. In: Isotope hydrology, 1983, IAEA, Vienna, pp 367–395Google Scholar
  55. Ousmane B, Galadima S, Moumouni A, Soumana I (2012) La qualité physico-chimique des eaux des aquifères discontinus du socle du Département de Téra (Liptako, Niger): impacts Sur les taux de desserte en eau potable des populations rurales au Sahel [Physico-chemical quality of water in basement discontinuous aquifers of the department of Tera (Liptako, Niger): impacts on the drinkable water supply rates in Sahel rural areas]. Africa Geosci Rev 19(1):1–16Google Scholar
  56. Panthou G, Vischel T, Lebel T (2014) Recent trends in the regime of extreme rainfall in the Central Sahel. Int J Climatol 34:3998–4006.  https://doi.org/10.1002/joc.3984 CrossRefGoogle Scholar
  57. Rana SS (1998) Application of directional filtering in lineament mapping for groundwater prospecting around Bhinmal: a semi arid part of Thar Desert. J Indian Soc Remote Sensing 26(1-2):35–44CrossRefGoogle Scholar
  58. Sagna P, Ndiaye O, Diop C, Niang AD, Sambou PC (2015) Les variations récentes du climat constatées au Sénégal sont-elles en phase avec les descriptions données par les scénarios du GIEC [Are recent climate variations observed in Senegal in conformity with the descriptions given by the IPCC scenarios]? Pollut Atmos 227:1–17Google Scholar
  59. Saravana KU, Suman S, Navada SV, Deodhar AS (2009) Environmental isotopes investigation on recharge processes and hydrodynamics of the coastal sedimentary aquifers of Tiruvadanai, Tamilnadu state. India J Hydrol 364:23–39CrossRefGoogle Scholar
  60. Scanlon BR, Keese KE, Flint AL, Flint LE, Gaye CB, Edmunds WM, Simmers I (2006) Global synthesis of groundwater recharge in semiarid and arid regions. Hydrol Process 20:3335–3370CrossRefGoogle Scholar
  61. Schiewede M, Duijnisveld WHM, Böttcher J (2005) Investigation of processes leading to nitrate enrichment in soils in the Kalahari region, Botswana. Phys Chem Earth 30:712–716CrossRefGoogle Scholar
  62. Séguis L, Cappelaere B, Milesi G, Peugeot C, Massuel S, Favreau G (2004) Simulated impacts of climate change and land-clearing on runoff from a small Sahelian catchment. Hydrol Process 18:3401–3413.  https://doi.org/10.1002/hyp.1503 CrossRefGoogle Scholar
  63. Solder JE, Stolp BJ, Heiweil VM, Susong DD (2016) Characterization of mean transit time at large springs in the upper Colorado River basin, USA: a tool for assessing groundwater discharge vulnerability. Hydrogeol J 24:2017–2033CrossRefGoogle Scholar
  64. Soumaila A, Konate M (2005) Caractérisation de la déformation dans la ceinture birimienne (paléoprotérozoïque) de Diagorou-Darbani (Liptako nigérien, Afrique de l’Ouest) [Characterization of the deformation in the Birimian belt (Paleoproterozoic) of the Diagorou-Darbani (Niger Liptako, West Africa)]. African Geosci Rev 12(3):161–178Google Scholar
  65. Stadler S, Osenbrück K, Suckow AO, Himmelsbach T, Hötzl H (2010) Groundwater flow regime, recharge and regional-scale solute transport in the semi-arid Kalahari of Botswana derived from isotope hydrology and hydrochemistry. J Hydrol 388:291–303CrossRefGoogle Scholar
  66. Taupin J-D, Gaultier G, Favreau G, Leduc C, Marlin C (2002) Variabilité isotopique des précipitations sahéliennes à différentes échelles de temps à Niamey (Niger) entre 1992 et 1999: implication climatique [Isotopic variability of Sahelian rainfall at different time steps in Niamey (Niger, 1992–1999): climatic implications]. C R Geosci 334:43–50CrossRefGoogle Scholar
  67. Taupin J-D, Gaultier G, Favreau G, Leduc C, Ramirez E (2003) Étude isotopique des précipitations sahéliennes à l’échelle de l’événement: relation avec les paramètres météorologiques et le type de précipitation [Isotopic study of Sahelian rainfall at the event scale: relationship with meteorological parameters and rainfall characteristics]. Hydrol Mediterranean Semiarid Regions 278:179–185Google Scholar
  68. Taylor C, Lambin E, Stephenne N, Harding R, Essery R (2002) The influence of land use change on climate in the Sahel. J Clim 15:3615–3629CrossRefGoogle Scholar
  69. Tremoy G, Vimeux F, Mayaki S, Souley I, Cattani O, Risi C, Favreau G, Oï M (2012) A 1-year long d18O record of water vapor in Niamey (Niger) reveals insightful atmospheric processes at different timescales. Geophys Res Lett 39:L08805, 5 pp.  https://doi.org/10.1029/2012GL051298
  70. Yaméogo OS (2008) Ressources en eau souterraine du centre urbain de Ouagadougou au Burkina Faso, qualité et vulnérabilité [Groundwater resources in the urban center of Ouagadougou in Burkina-Faso, quality and vulnerability]. PhD Thesis, University of Avignon and the Vaucluse region, France, Univ. Ouagadougou, Burkina Faso, 245 ppGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  • Maman Sani Abdou Babaye
    • 1
    • 2
    Email author
  • Philippe Orban
    • 3
  • Boureima Ousmane
    • 1
  • Guillaume Favreau
    • 4
    • 5
  • Serge Brouyère
    • 3
  • Alain Dassargues
    • 3
  1. 1.Faculté des Sciences et Techniques, Département de GéologieUniversité Abdou MoumouniNiameyNiger
  2. 2.Faculté des Sciences et Techniques, UMR SERMUG, Département de GéologieUniversité Dan Dicko Dankoulodo de MaradiMaradiNiger
  3. 3.Faculté des Sciences Appliquées, DépartementArGEnCo, Laboratoire d’Hydrogéologie et Géologie de l’EnvironnementUniversité de LiègeLiègeBelgium
  4. 4.IRD, UMR HydroSciencesUniversité de MontpellierMontpellierFrance
  5. 5.Institut de Recherche pour le Développement (IRD), CNRS, Institute of Engineering Univ. Grenoble Alpes (G-INP), Institut des Géosciences de l’Environnement (IGE)University Grenoble AlpesGrenobleFrance

Personalised recommendations