Introduction

There are three commonly addressed questions in hydrogeological investigations: (1) Where does the water come from (recharge area)? (2) Where does the water go? (3) What is the depth and lateral extent of groundwater circulation? These questions refer to the spatial scale of groundwater flow, in which a groundwater flow system can be divided conceptually into “local” and “regional” groundwater flows (e.g. Toth 1962, 1963; Freeze and Witherspoon 1967). Local flow is limited within a single basin enclosed by topographic high points and circulates to relatively shallow depths in the subsurface. Regional flow circulates to greater depths and is not necessarily limited within a single hydrologic basin. There are three methods to determine the groundwater flow in a wide area: measuring the fluid potential, numerical simulation and use of tracers. The tracers for groundwater surveys are classified based on methods of detection into six kinds; colorimetry, photometry, mass spectrograph, electric conductivity, chemical components and temperature (Todd 1959).

Cartwright (1970) and Sakura (1978) used temperature data to understand groundwater flow in a basin. Generally, the normal geothermal gradient of a subsurface regime is affected by groundwater flow and changes in surface temperature. A theoretical analysis of a temperature field distorted by topographically driven groundwater flow was presented by Domenico and Palciauskas (1973). Their results show that groundwater temperature increases with depth under the influence of both heat conduction and advection. At the same altitude, subsurface temperature in the recharge area is lower under the condition of groundwater flow than under no flow conditions. In the discharge area, temperature is higher with groundwater flow than under no flow conditions.

Stable isotopes of δ18O and δD have been used extensively for the past several decades as tracers. As they form the water molecule itself, they can be considered ideal tracers for the physical movement of water. That is, for most groundwater in the “normal” temperature range, there are no modifications due to interactions between the aqueous phase and solid minerals. The stable isotopes of δ18O and δD give information about recharge and discharge processes, and can be used to estimate recharge areas and mixing fractions between aquifers and between surface water and groundwater, as well as to provide a window into past changes in recharge or climatic condition during the recharge processes (Tyler et al. 2000).

As water moves into the ground it begins to record information on the history of its recharge source and properties, mainly from rainfall solutes as well as isotopic ratios in the water molecule. The chemical components and distinctive characteristics of groundwater are mainly established in the unsaturated zone. In the saturated zone, the geochemical evaluation, though less intense than in the soil and unsaturated zones, follows progressive changes in water-quality towards areas of discharge. These processes are time dependant and the chemical changes as well as isotopic variations may be used to provide information on water flow paths (Herczeg and Edmunds 2000). In this study temperature, stable isotopes and water-quality were used for tracing the groundwater flow system in the Nagaoka Plain. Flow analysis is primarily based on the concept of hierarchical groundwater flow systems (Toth 1962, 1963).

Hydrogeological Setting of the Nagaoka Area

The Nagaoka area is located in the south of the Niigata Plain along the cost of the Japan Sea, Honshu, Japan (Fig. 1). The main geologic structure of the Nagaoka area was formed by a series of crustal movements which occurred during the Late Pliocene and Pleistocene. The surrounding hills (up to 2000 m a.s.l.) are composed of thick sedimentary rocks of the Plio-Pleistocene Period, which form the basement of the main aquifers in the plain. While the plain surface is covered mainly by urban and paddy field areas, the surrounding hills are mainly covered by forests (Fig. 2). According to borehole records and aquifer test data, the major aquifer consists of four formations (Fig. 3) in this groundwater basin. The first and second formations are gravel rich alluvial deposits with a total thickness of about 20–40 m. The third and the fourth formations are silt containing deposits. The plain surface is covered with a clayey layer. The northern part of the area and tributary streams have a thick clayey layer. Main hydraulic conductivities of the formations which are identified as layers I, II, III and IV are 0.12, 0.092, 0.022 and 0.018 cm/s respectively in descending order (the Nagaoka Construction Office 1975).

Fig. 1
figure 1

Map showing the location of the Nagaoka area and the watershed elevations in meters

Full size image
Fig. 2
figure 2

Location and land use map of the Nagaoka area; cross-section lines A-A’, B-B’, and C-C’ and sampling locations are also shown. Note, filled circles are observation wells, crosses are pumping wells, river samples, and filled triangles refer to the hot springs locations

Full size image
Fig. 3
figure 3

Two cross-sections. a Location of the cross-section lines. b and c Show the vertical lithological succession of the main aquifer along the lines a-a’ and b-b’

Full size image

As mentioned by Taniguchi (1986), the mean annual precipitation is 2,654 mm/year; about 40–50% of the precipitation falls as snow from November to March. The annual evapo-transpiriation calculated by Penman’s method is 707 mm/year. This area is the first place where groundwater was used for melting snow dating back about 40 years. The depth of the pumping wells range between 5 to 250 m, but mostly between 20 to 60 m.

Taniguchi (1986), stated that the Shinano River is naturally a gaining stream fed from groundwater all through the reach in the Nagaoka area, except during the winter season when the water table at the center of the city drops to a level lower than the level of the Shinano River owing to the heavy use of groundwater for melting snow. Decline of the water table induces artificial recharge from the river into the groundwater (Kayane et al. 1985). When this practice started, the recovery of the water table was very rapid in April due to cession of pumping. The water table now does not recover fully until July or August.

The relation between river water and groundwater has been changed due to groundwater use. The Shinano River has become a recharging source for the groundwater in the urban area most of the year, which may affect the subsurface environment of the urban area. The main groundwater flow direction is from south to north as shown from the horizontal distribution of the hydraulic heads in the plan area (Fig. 4). The vertical two-dimensional distribution of the hydraulic heads along the lines A-A’ and B-B’ are shown in Fig. 5a, b. In December the water flows smoothly from south to north with a gentle slope. In March the water table drops with steep slopes under the urban area, and groundwater flows from different directions toward the urban area. After March, the water table starts to recover and as shown in the diagram it does not recover till June and is not expected to reach the normal level until July or August.

Fig. 4
figure 4

Contour map showing the water table distribution in Nagaoka area (Groundwater Map 1998)

Full size image
Fig. 5
figure 5

Vertical two-dimensional distribution of the hydraulic head during June, December (2000) and March (2001) along the lines A-A’ ( a) and B-B’ ( b). Note, filled circle represents the screen center

Full size image

Methodology

Subsurface thermal measurements generally were made in observation wells that were assumed to be in equilibrated thermal state between the water and the surrounding solid material. Temperature measurements in this study were carried out in 30 observation wells (20–115 m depth). The equipment used for the measurements was a digital thermistor thermometer (resolution of 0.01 °C) attached to a cable of 300 m. Data were recorded from the water table to the bottom of the hole at 2-m intervals.

Water for chemical and isotope analysis was sampled from most of the observation and pumping wells (25–195 m depth). Water table, electrical conductivity as well as the pH of the groundwater were measured in situ. Anions such as NO3, Cl and SO4 were measured by ion chromatography and cations such as Na, K, Ca, Mg, and SiO2 were measured by ICP. HCO3 (alkalinity) was measured by 0.02 N H2SO4 titration method. Stable isotopes δ18O and δD were measured by mass spectrometer.

Results

Temperature

Taniguchi (1986) studied the subsurface temperature in the Nagaoka area. His work can be summarized by the following: (1) the infiltrated snowmelt water makes the soil temperature decrease at all depths in the capillary water zone in the daytime, (2) seasonal changes in temperature-depth profiles were classified into four characteristic types—recharge, discharge, advection, and pumping, and (3) the infiltration rate in the recharge area is about two times as great as that of the discharge area as shown in Fig. 6. In this study the values >0.7, 0.6–0.7 and <0.5 m/day are considered as high, intermediate and low infiltration rates, respectively.

Fig. 6
figure 6

Distribution of the infiltration rate for 156 mm rainfall in Nagaoka area (Taniguchi 1986)

Full size image

A two-dimensional vertical subsurface temperature distribution of December data along the south–north main groundwater flow direction A-A’ line (Fig. 2) is shown in Fig. 7a. In this figure, the main groundwater flow system is estimated as follows: the groundwater recharges from the south and discharges north to Nagaoka City where the subsurface temperature is higher than in the recharge area. The existence of a warm zone higher than 16.0 °C under the main urban area was explained by Salem et al. (2004) as an effect of urban warming. The effect of Shinano River on the subsurface temperature is recognized clearly from Fig. 7b. A cold zone less than 12.0 °C appears under the Shinano River as a result of induced recharge from the Shinano River due to the 40-year-groundwater pumping history in the urban area.

Fig. 7
figure 7

Vertical two-dimensional distribution of December (2000) temperature data along the lines A-A’ ( a), B-B’ ( b) and C-C’ ( c), respectively. Arrows show the estimated groundwater flow direction. Observation wells are shown by numbers and temperatures units are in °C

Full size image

Figure 7c shows an east–west subsurface temperature distribution along the line C-C’. In general, it is estimated that the groundwater in this area flows from both the east and west to the central part of the plain and discharges into Shinano River. The shallow warm zone under the area of well 33 could be explained as an effect of the urban warming. The area of well 25 shows an independent flow system where the deep groundwater recharged in the eastern mountains is assumed to discharge. The same phenomenon is also noticed in the one-dimensional temperature profile of well 24 (Fig. 8). Oki et al. (1996) explained the mechanism of such discharge process as a discharge along a tectonically active zone. This explanation could be confirmed by existence of many hot springs along the contact between the high land and low land in the eastern part of the studied area (Fig. 2).

Fig. 8
figure 8

A temperature-depth profile of well 24 shows high subsurface temperature

Full size image

Stable Isotopes

Hydrogen and oxygen stable isotopes are ideal tracers for estimating the recharge areas and flow path of the groundwater, because they compose the water molecules and are sensitive to physical processes such as mixing and evaporation (Coplen 1993). Mizutani et al. (2001) studied the isotopic relation between river water and groundwater infiltrated from a paddy field area in the Toyama Plain (south of the Nagaoka area). They found that a paddy field irrigated from the main river flowing in fan deposits showed isotopic fractionations of δD=12.3‰ and δ18O=2.6‰ on average. This result was used in this study because both the Toyama and Nagaoka areas have similar weather conditions. Shinano River water and its tributary waters were sampled three times in March, October and December. The main isotopic composition of the river water was calculated and the isotopic composition of the groundwater infiltrated from paddy fields was estimated from the previous relation.

The stable isotope data of δD and δ18O‰ (Table 1) are variable. From δ-diagram (Fig. 9) it was found that the surface water and groundwater of the Nagaoka area are a mixture of precipitation that fell during different seasons of the year, because all the samples are located between the summer and winter local meteoric lines (Fig. 9). As shown in the δ-diagram, the plotted samples can be classified clearly into four main groups: A, B, C and D. Group A represents the deep groundwater, which has the lightest isotope composition indicating a high altitude recharge area. Group B represents the intermediate groundwater flow system as well as three shallow samples of wells 15, 26 and 11. The aerial distribution of the later three samples is local and limited to the head of the alluvial fan (i.e., the area with high infiltration rate). This groundwater infiltrated mostly from precipitation. Groundwater samples of group D represent the shallow groundwater that infiltrated mostly from the paddy-field water. Group C samples generally represent the shallow groundwater of the area where the infiltration rate ranges from intermediate to low including the urban area. Group C also plots in a position intermediately between B and D groups.

Table 1 Isotopic compositions of the analysed water samples
Full size table
Fig. 9
figure 9

δ-diagram showing four types of groundwater in the Nagaoka Plain

Full size image

The two-dimensional vertical subsurface distribution of oxygen isotopes along the lines A-A’ and B-B’ are illustrated in Fig. 10a, b. The water table during December and March is also shown. The oxygen isotopic distribution shows shallow, intermediate and deep flow systems. The shallow system has its local component in the area of wells 15, 26 and 11 (group B in Fig. 9). Combining the water table data and the aerial distribution of the infiltration rate with the subsurface distribution of oxygen isotopes is useful for explaining the distribution of the isotopic composition in the shallow zones. Figure 10b shows that the sites of wells 10, 35 and 34 are mostly recharged by the paddy field water (group D in Fig. 9) as they have the heaviest oxygen isotopic composition in the area. It is also shown that the river water in the site of well 5 has recharged the groundwater under the effect of pumping (as predicted by the existence of a cold temperature zone under the Shinano River). The shallow groundwater samples of the urban area (ex. 2, 3, 7, 8 and 102) and some other sites on cultivated land (ex. 107 and 113) are located in an area of intermediate and low infiltration rates. Such infiltration rates may lead to isotopic fractionations and/or mixing between the precipitation water and the paddy-field water in the cultivated land during infiltration. The intermediate flow system has an oxygen isotopic composition of δ18O‰=−9.0 to −9.3. The deep flow system has the lightest oxygen isotopic composition with δ18O‰=−9.5 to −9.7.

Fig. 10
figure 10

Vertical two-dimensional distribution of oxygen stable isotope data in groundwater. Arrows show the flow direction

Full size image

Hydrochemistry

The chemical composition of the shallow groundwater in the Nagaoka plain is variable (Tables 2 and 3). The electric conductivity (EC) of the shallow groundwater below the urban area ranged from about 37.3 to 498 μs/cm in December, but it reached 1,240 μs/cm (ex. well 4) in March when the groundwater is heavily pumped for snowmelt. EC of the shallow groundwater below the cultivated land ranged from 120 to 420 μs/cm. The deep groundwater samples have the lowest electric conductivity (lower than 150 μs/cm). Generally the electric conductivity decreases downward, and in shallow zones it reaches its maximum under the urban area especially in the wintertime.

Table 2 Chemical analysis of December (2000) water samples (symbols are shown in Fig. 2). Electric conductivity ( EC) values are in (μS/cm) and concentrations are in mg/L
Full size table
Table 3 Chemical analysis of March (2001) water samples (sample’s symbols are shown in Fig. 2). Electric conductivity ( EC) values are in (μS/cm) and concentrations are in mg/L
Full size table

The shallow groundwater of the main part of the urban area, moderately deep and deep groundwater, is free from NO3. Nitrates exist only in the shallow groundwater that underlies the cultivated land and some wells in the surrounding urban area.

Silicate was measured as the concentration of SiO2. In the study area, the shallow groundwater of the urban area has low SiO2 concentration (from 0.70 to 2.35 mg/L). The moderately deep groundwater and the deep groundwater show high concentration (from 22 to 30 mg/L). The shallow groundwater of the cultivated area has an intermediate concentration of SiO2 (from 10 to 15 mg/L). Such distribution probably depends on the water–rock interaction. A piper diagram of December data (Fig. 11a) shows that Nagaoka groundwater can be classified into three main types, sulfate-chloride, carbonate, and sodium carbonate groundwater. The first type (sulfate-chloride type) represents the shallow groundwater underlying the urban area (samples 3, 2, 4, 6, 7, 8, 102, P06). The second type (carbonate water) represents the shallow groundwater underlying the cultivated land, moderately deep and the deep groundwater. Sodium carbonate water is represented by well 5 water, which infiltrated from the Shinano River under the effect of pumping. The river water and the river type groundwater are also represented. The river type groundwater is represented by samples 20, 22, P10 and P11. The first two samples are from wells in the western side of the Shinano River and the later two are from wells in the recharge area. March data (Fig. 11b) shows an increase in sulfate concentration in the shallow groundwater of the urban area (note, deep groundwater samples are not plotted in this diagram because they have no seasonal changes).

Fig. 11
figure 11

Piper diagrams show different types of groundwater in the studied area. a Shows the December (2000) data and b shows the March (2001) data. Note, the symbols are as following, filled circles refers to shallow groundwater of the urban area, open circles are river type groundwater, filled diamonds are well 5 water infiltrated from river water, open squares are deep groundwater, and open triangles are river water

Full size image

Figure 12 is a Stiff diagram showing the vertical two-dimensional distribution of the groundwater quality data of December 2000 along lines A-A’ and B-B’ shown in Fig. 12a, b. Three groundwater flow systems are identified: shallow, intermediate and deep systems. The shallow groundwater and intermediate flow systems were generally composed of CaHCO3 type water. The shallow groundwater type shifted to CaSO4 type in March under the urban area (compare between Fig. 12b, c). The Intermediate groundwater shifted to CaCl2 type under the central part of the urban area in its upper zone (sample P06). The deep groundwater was composed of Mg–Ca–NaHCO3 type with about equal concentrations of Mg, Ca, and Na. It also had low chloride concentration and no sulfates. Water of well 5, which is recharged by Shinano River, had a unique chemical type (NaHCO3) that had very low dissolved salts and was quite different from the surrounding groundwater. It is shown from the chemical data that the groundwater of the urban area was highly altered especially in the winter season and the sequence of the groundwater flow system under the urban area was quite different from that of the rest of the plain.

Fig. 12
figure 12

The vertical two-dimensional distribution of the groundwater quality along the lines A-A’ ( a) and B-B’ ( b) respectively. ( c) shows the increase in sulfate concentration in shallow groundwater samples under the urban area during March (2001)

Full size image

Discussion

Using multi-tracers is a good tool for formulating a complete concept of the groundwater flow system in an area. In this paper, subsurface temperature, stable isotopes and water quality were used to define the groundwater flow path. The Nagaoka area topographically consists of (1) a plain area covered by two main types of land use, urban areas and the paddy field areas, and (2) the surrounding forested mountain areas. The groundwater flow systems in the plain area have two main flow directions, south–north (Figs. 7a, 10a and 12a) and east and west to the central part of the plain of the Shinano River (Fig. 7c). Depending on the stable isotopes and the water quality, the south–north flow direction could be classified into three main flow systems: shallow, intermediate and deep. There is another independent flow system discharging along the contact between the high land and plain area. The later flow system could be recognized by the existence of a high temperature zone in Figs. 7c and 8. Wells 25 and 24 can be considered as a part of this system from a line passing from south to north and connecting between the hot springs. Oki et al. (1996) considered such lines as marking a buried active fault where the epicenters of many historical earthquakes occurred. It should be noted that this flow system is a very deep groundwater discharge along a tectonically active zone and quite different from the above mentioned three flow systems of the plain area.

Generally, the groundwater in the studied area is a mixture of the different types of precipitation through the year (Fig. 9). In contrast to the intermediate and deep flow systems, the shallow flow system has been affected by many surface phenomena such as (1) the land use, (2) location of Shinano River in the center of the plain west to the main city area where the groundwater has been heavily pumped for snowmelt for the last 40 years, and (3) the aerial distribution of the infiltration rate in the plain area (Fig. 6). A wide area covered by paddy fields (Fig. 2) and the aerial distribution of infiltration rate (Fig. 6) control recharge to the shallow groundwater in the studied area. Recharge by the paddy field waters was recognized as group D (Fig. 9) and it had the heaviest stable isotopic composition due to extensive evaporation before infiltration. In the area at the top of the alluvium fan, which has a high infiltration rate, the shallow groundwater was mostly recharged by precipitation. The later plots as group D in δ diagram (Fig. 9) and forms a short flow path as in Fig. 10a. As shown in Figs. 9 and 10, the shallow groundwater in the areas where the infiltration rate ranged from intermediate to low, shows intermediate isotopic composition between both the groundwater recharged from precipitation and that recharged from the paddy fields. Thus, the isotopic fractionation and/or mixing between precipitation waters and the paddy field water (irrigated by river water) may have occurred before infiltration.

Chemically, the shallow groundwater underlying the paddy fields shows a different trend compared to that underlying the urban areas. The former is a CaHCO3 type water with some NO3 and a considerable concentration of SiO2 (10–15 mg/L). Respiration or oxidation of organic matter is the major source of CO2 in the soil. Considering soil water as pure water in which CO2 dissolves, the system can be described as follows (Appelo and Postma 1994):

$$ {\text{CO}}_{{\text{2}}} + {\text{H}}_{2} {\text{O}} \leftrightarrow {\text{H}}_{{\text{2}}} {\text{CO}}_{{\text{3}}} + {\text{H}}^{ + } \leftrightarrow {\text{HCO}}_{{\text{3}}} ^{ - } + {\text{H}}^{ + } \leftrightarrow {\text{CO}}_{3} + {\text{H}}_{2} . $$
(1)

The origin of carbonates in such shallow groundwater can be explained by the same reaction because the soil probably is rich in organic matter. The main cause for the existence of nitrates in this water is the application of fertilizers during the time of rice cultivation (Appelo and Postma 1994). SiO2 in this groundwater indicates that water has reacted with silica bearing sediments or rocks, possibly including the shallow aquifer material and upper clay layer of the unsaturated zone.

Pumping and urbanization have affected the groundwater of both the shallow and the intermediate flow systems underlying the urban area. In the shallow zones three main phenomena were recognized including (1) seasonal fluctuation of sulfate concentration, (2) temperature, isotopic and chemical composition of well 5 is quite different compared with the surroundings, and (3) high chloride concentration of well P06 under the main city area. The oxidation of pyrite and other sulfides by exposing saturated sediments to the atmosphere by lowering the groundwater table, has a large environmental impact and plays a key role as a source of sulfate and iron in the groundwater as well as a source of heavy minerals in the environment (Appelo and Postma 1994). This process occurs as in the following reaction:

$$ {{\rm{FeS}}_{2} + 15/4{\rm{O}}_{2} + 7/2\;{\rm{H}}_{2} {\rm{O}} \to {\rm{Fe}}{\left( {{\rm{OH}}} \right)}_{3} + 2{\rm{SO}}_{4} ^{{2 - }} + 4{\rm{H}}^{ + } .} $$
(2)

The same probably happened in the studied area during March 2001 when the water table was lowered by heavy pumping. On the other hand, after March, the groundwater table started to recover and reached its steady state in August. While the water table is close to the surface, downward transport of DOC (dissolved organic carbon) may deplete oxygen in the aquifer (Starr 1988; Starr and Gillham 1989) which may lead to reduction of sulfates and formation of carbonates in the shallow zones under the urban area. This process is described by Appelo and Postma (1994) as in the following reaction:

$$ {{\rm{SO}}_{{\rm{4}}} ^{{{\rm{2}} - }} + {\rm{2CH}}_{{\rm{2}}} {\rm{O}} \to {\rm{H}}_{{\rm{2}}} {\rm{S}} + {\rm{2HCO}}_{{\rm{3}}} ^{{^{ - } }} .} $$
(3)

Salem et al. (2004) studied the subsurface temperature history of well 5. They suggested that the Shinano River in that area has been changed from a gaining to a losing stream. Therefore, the presence of a cold zone (Fig. 7b) and the difference in the isotopic and chemical composition of the groundwater compared to the surrounding (Figs. 10b and 12b) area of well 5 can be related to induced recharge from the river to the groundwater under the effect of long term pumping. The front of the groundwater induced from Shinano River is characterized by NaHCO3 type (Figs. 11 and 12b). Formation of this chemical type is probably related to ion exchange between the induced river water and the groundwater of high electric conductivity under the urban area. This process is called freshening (Appelo and Postma 1994) and it is described as follows:

$$\raise0.5ex\hbox{$\scriptstyle {\text{1}}$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle {\text{2}}$}{\text{Ca}}^{{{\text{2}} + }} + {\text{Na}} - {\text{X}} \to \raise0.5ex\hbox{$\scriptstyle {\text{1}}$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle {\text{2}}$}{\text{CaX}}_{2} + {\text{Na}}^{ + } $$
(4)

where Ca2+ is taken up from the water, in return for Na+ with NaHCO3 type water as a result.

In winter, NaCl is used in Nagaoka City for de-icing along the main streets. Some of the snowmelt polluted with NaCl may infiltrate to the groundwater. Under the central part of the urban area, the dominant anions in the shallow zones, where the circulation of the groundwater is high, are SO4 and/or HCO3, depending on the seasonal variations with less concentration of Cl. In contrast, in the moderately deep groundwater, the dominant anion is Cl (well P06). This sequence was detected by Domenico (1972) which he related to the downward decreases in groundwater circulation. As it is shown in Fig. 12a, the groundwater of well P06 is CaCl2 type. Formation of this chemical type is related to an ion exchange process (Appelo and Postma 1994) as follows:

$$\raise0.5ex\hbox{$\scriptstyle 1$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle 2$}{\text{Na}}^{ + } + \raise0.5ex\hbox{$\scriptstyle 1$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle 2$}{\text{Ca - X}}_{2} \to {\text{Na - X}} + \raise0.5ex\hbox{$\scriptstyle 1$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle 2$}{\text{Ca}}^{{2 + }} $$
(5)

where X indicates the exchanger. Then the exchanger takes up sodium and Ca is released. The water-quality thus changes from NaCl to CaCl2 type water.

The Silica concentration of the shallow groundwater underlying the urban area is the lowest among the Nagaoka Plain groundwaters (from 0.70 to 2.35 mg/L) This low concentration is because, (1) under the effect of pumping there is not enough time for interaction between infiltrated water and silicate minerals of the host rocks (Roy 1954 and Krauskoph 1956), and (2) there is precipitation of insoluble quartz and chalcedony (White et al. 1956).

The intermediate groundwater flow system is composed of CaHCO3 type water and its stable isotopic composition is intermediate between the shallow and deeper flow systems (from –9.0 to –9.3). The high carbonate concentration may be related to oxidation of the organic matter in the soil zones of the recharge area as shown in Eq. (1). Unlike the shallow and the intermediate flow systems, the oxygen isotopic composition of the deep system is the lightest (−9.5 to −9.7) and it contains about equal concentrations of Na, Ca, Mg and the main anion is the CO3. These differences probably relate to the high altitude of the recharge area and the aquifer rock composition, respectively. Both the intermediate and the deep systems have higher silicate concentrations (about 26 mg/L on average) compared to the shallow one. In Fig. 9, isotopic composition of group A samples are the nearest to the river water, possibly because the catchment areas of both river and deep groundwater are the same.

Conclusion

Using multi-tracers is a good reconnaissance tool for estimating groundwater flow systems in an area. In this paper, subsurface temperature, stable isotopes and water quality were used to determine the groundwater flow path. The main findings can be summarized as follows:

  1. 1.

    The groundwater flow system in the Nagaoka Plain area has two main flow directions, a south–north direction and the flow from both east and west to the central part of the plain to the Shinano River

  2. 2.

    There is another independent flow system discharging along the contact between the high land and plain area. The later flow system can be recognized by the existence of a high temperature zone in wells 25 and 24.

  3. 3.

    In general, the groundwater in the studied area is a mixture between the different types of precipitation through the year

  4. 4.

    The flow system along the south–north flow direction can be classified into three main flow systems including shallow, intermediate and deep

  5. 5.

    The shallow system is highly affected by the land cover, urban pollution and heavy pumping. Long-term pumping in the urban area changed the Shinano River from a gaining to a losing stream.