The stable isotopes of oxygen-18 and deuterium were utilized to identify different water sources mixing in the soil layer that recharged to groundwater of a paddy field during the growth season in 2014. Based on the measurements of deuterium and oxygen-18 in soil water, rainwater and groundwater in the paddy field of Chianan Plain in Southwest Taiwan in the wet season was collected, and the relationship between δD and δ18O in soil water and groundwater recharge was investigated in this study. The soil water in the paddy field of Chianan plain was collected with suction lysimeters for the identification of different sources of subsurface flow. The isotopic compositions of hydrogen and oxygen in different water bodies were used to evaluate the groundwater recharge sources in the paddy field. The intensity of rainfall and evaporation influenced the saturation conditions of pore water present in the soil layer. In general, transiently intense rainfall tended to be difficult in recharging groundwater. The results show the use of stable isotopes in water bodies can be used to monitor or identify the source/sink of soil layer and their respective contributions to groundwater recharge. In this research, the isotopic compositions of hydrogen and oxygen in the soil water at different depths before and after event water were identified. The top soil layer (<30 cm depth) in the presence of the isotope depleted soil water was probably subjected to evaporation. The soil water has more depleted isotope composition that was observed in shallow soil layers. More depleted fraction of isotopes in groundwater was similar to rainwater, which suggested that the groundwater was primarily from rainfall. In addition, the isotopic compositions of hydrogen and oxygen in groundwater slightly deviated from the local meteoric water line in southern Taiwan.

  • Analyze the groundwater infiltration mechanism of rainfall in the Chianan Plain of Taiwan.

  • Hydrogen and oxygen isotopes as a tracer in hydrogeological research.

  • Contribution of different rainfall types to groundwater recharge.

  • Reveal the time required for rainfall to infiltrate the underground aquifer.

  • As a reference for the development and utilization of groundwater resources.

Different water sectors in the soil environment play a significant role in the hydrological cycle. For example, the rainwater is the utmost source of surface water and groundwater, isotopic characteristics of rainwater are thus inherently helpful for the hydrological studies in Taiwan (Peng et al. 2010). Rainfall on the ground surface can be intercepted by plant leaves and subsequently be evaporated. Infiltration of rainfall in the complicated process can be affected by soil structure, texture, moisture and extent of heterogeneity (Gazis & Feng 2004). Rainwater infiltrates directly from surface drainage through pore spaces in soil and the amount of water was held against gravity on the grains of the soil by adsorption and capillary forces (Dincer et al. 1974; Gvirtzman et al. 1986). The process of infiltration has been widely studied.

Several methods are available to distinguish different water sources using hydrometer-based approaches, chemical tracers, modeling, and stable isotope applications (Orlowski et al. 2014). Many studies elaborated different sources of water partitioning commonly based on interpretations of isotope dynamics in the small watershed or field plot scale (Shanley et al. 2002). The isotopic compositions of hydrogen and oxygen in groundwater that are recharged by direct infiltration to the unsaturated zone follow the incident precipitation rather closely (Gat & Tzur 1967). On a local region, the atmospheric conditions setting of the environment can significantly influence the isotopic ratios of meteoric waters. The deuterium-excess parameter defined by Dansgaard (1964) as d = δD − 8 *δ18O has been correlated mainly with the environmental conditions existing in the source area of the water vapor (Gat & Carmi 1970).

Previous studies indicated that soil water can form piston flow, in which water from more recent rainfall events pushes the older soil water to flow downward (Komor & Emerson 1994). The water can be divided into event (new) water and pre-event (old) water. Event water is the water from wet precipitation that enters the subsurface during the rain event, whereas pre-event water is the water in a vadose zone and groundwater in the saturated zone (Zhao et al. 2013). In addition, the isotope values were relatively light in the plain area but gradually enriched in the foothills belt. Average isotope values in groundwater of the plain area match closely with those in precipitation (Sengupta et al. 2014). Soil water is taken up by plants essentially unfractionated and this feature is used extensively to characterize the source of water for plant and transpiration (Dawson & Ehleringer 1991). The soil water in the top soil layer had the shortest residence time and most of the pre-event (old) water was dispelled in a storm (Zhao et al. 2013). The stable isotopic composition in the shallow soil layer is dominated by individual rainfall events (Vogel et al. 2010). Generally, isotopic composition of hydrogen and oxygen in soil water was much less variable than that of rainfall, indicating isotopic mixing between waters of different rainfall events, whereas the isotopic δD and δ18O values of summer rainfall decreased with increasing depth, and soil at 50 cm depth could only receive water from large storms (Tang & Feng 2001). A number of hydrologic studies indicate that the isotope analysis can identify the storm hydrograph that is dominated by pre-event water (Hooper et al. 1990; Laudon & Slaymaker 1997; Shanley et al. 2002). The evident spatial heterogeneity of isotopic compositions in different sources of water revealed a complexity problem and limited deep insights into hydrological processes (Kendall et al. 2001). Furthermore, altitude and temperature effects are very prominent in the stable isotope signals (Peng et al. 2010). The natural isotope tracers cannot be controlled and vary seasonally over the years (Bengtsson et al. 1987). However, little is known about the mixing of infiltrating rainfall water and soil water or about its recharge contribution to shallow groundwater. The stable isotopes of hydrogen and oxygen can be used to assess these interactions and mixing in the unsaturated zone. This study investigated the isotopic compositions (δD and δ18O) in soil water of a paddy field where the isotopic composition of different sources of water can change significantly from event water to pre-event water. The stable isotopes are also used to determine water affected by evaporation and to identify evaporation fronts in a soil profile (Komor & Emerson 1994).

The lithology in the unsaturated zone in the Hsuechia experimental site of this study consists of homogenous fine sand in top soil and interlayered sand and silt in deep soil. The site has been planted with rice (Oryza sativa L.) in the wet season and corn in the dry season and the experiment plots irrigated according to local farming practices.

The present study investigated soil water movements in the paddy field and till soil using hydrogen and oxygen isotopes. The main goals in this study were to investigate the isotopic characteristics of the As-rich paddy soil. Consequently, the objectives of the study were to (i) investigate the isotopic compositions of hydrogen and oxygen in different depth profiles of soil water before and after rainfalls on the paddy field; (ii) decipher the different sources of water that mixed in soil layer; and (iii) clarify the δD and δ18O between groundwater and rainwater at different depths under the conditions of different soil water contents. Thus, this study was to detect different water sources in the unsaturated zone using hydrogen and oxygen isotopes.

Study site

The Hsuechia experimental site (23 °12′4.02″N, 120 °10′50.7″E) (Figure 1(a)) in a farm field (about 35 km from National Cheng Kung University campus) and a property of the Chianan Irrigation Association, is located in a sub-tropical zone with warm and humid weather in the wet season (April – September), in contrast to a cool temperature and low rainfall in the dry season (October – March). The area has an annual mean air temperature of 25 °C, with monthly mean air temperatures of 21 °C in January and 28 °C in July (Table 2). Within this study area, about 80% of the annual rainfall (in the range of 1,800–2,400 mm) is concentrated in the period of May–October. The dug soil profile (Figure 1(b)) shows that the soil is mostly composed of alluvium, with average grain size distribution of 13.2% clay, 72.5% sand and 14.2% silt (Chou et al. 2014). The subsurface soil water was monitored in the paddy field during two storms (Typhoons Matmo and Fung-Wong) on 23 July 2014 and 20 September 2014. The experimental field is about 100 m long and 40 m wide. Periodic irrigation with groundwater was applied during the growing season and agricultural practices (e.g., pesticide spray, weeding).

Figure 1

(a) Location of the experimental site at Hsuechia, Tainan City, in southwestern Taiwan, and (b) the soil profile dug in the studied site with the lithologic description in Table 1.

Figure 1

(a) Location of the experimental site at Hsuechia, Tainan City, in southwestern Taiwan, and (b) the soil profile dug in the studied site with the lithologic description in Table 1.

Close modal
Table 1

Basic physico-chemical and chemical properties of the soil at different depths of soil profile (Figure 1(b)) dug in the studied site

Depth (cm)Porosity (%)Water content (%)Organic carbon (%)CEC (cmol kg−1)Sand (%)Silt (%)Clay (%)
0–20 52.5 ± 1.4 99.3 1.3 ± 0.11 8.2 ± 0.9 45.6 ± 1.8 38.7 ± 1.6 15.6 ± 1.1 
20–40 51.2 ± 1.2 97.2 1.4 ± 0.21 7.3 ± 0.6 46.3 ± 1.9 20.5 ± 1.3 15.7 ± 1.2 
40–60 49.4 ± 1.1 91.1 1.2 ± 0.11 6.8 ± 0.5 69.9 ± 3.3 14.9 ± 1.2 12.6 ± 0.9 
60–80 41.7 ± 1.2 90.2 1.2 ± 0.11 7.6 ± 0.7 75.4 ± 3.2 10.8 ± 1.1 10.8 ± 0.9 
80–90 42.9 ± 1.3 89.6 1.2 ± 0.12 12.8 ± 1.2 86.6 ± 3.9 6.3 ± 0.8 7.3 ± 0.4 
Depth (cm)Porosity (%)Water content (%)Organic carbon (%)CEC (cmol kg−1)Sand (%)Silt (%)Clay (%)
0–20 52.5 ± 1.4 99.3 1.3 ± 0.11 8.2 ± 0.9 45.6 ± 1.8 38.7 ± 1.6 15.6 ± 1.1 
20–40 51.2 ± 1.2 97.2 1.4 ± 0.21 7.3 ± 0.6 46.3 ± 1.9 20.5 ± 1.3 15.7 ± 1.2 
40–60 49.4 ± 1.1 91.1 1.2 ± 0.11 6.8 ± 0.5 69.9 ± 3.3 14.9 ± 1.2 12.6 ± 0.9 
60–80 41.7 ± 1.2 90.2 1.2 ± 0.11 7.6 ± 0.7 75.4 ± 3.2 10.8 ± 1.1 10.8 ± 0.9 
80–90 42.9 ± 1.3 89.6 1.2 ± 0.12 12.8 ± 1.2 86.6 ± 3.9 6.3 ± 0.8 7.3 ± 0.4 

Sampling and site measurements

Rainwater was collected during rainfalls in 2014 (Table 3) with a high density polyethylene bottle connected with a 200-mm diameter glass funnel. The amount of rainfall was measured with a tipping bucket rain gauge (0.5 mm resolution in reading). During the course of this study, samples of groundwater (the groundwater level is approximately 120 cm deep) and soil water were collected periodically. Soil water was extracted using a low tension lysimeter. Soil water samples were collected at 5 different depths (0–20, 20–40, 40–60, 60–80, 80–90 cm) by low tension (a maximum tension of 85 kPa was applied) porous cup connected with a vacuum hand pump (Soil moisture Equipment Corp., CA, USA).

Isotopic analysis

Two-component (δD and δ18O) isotopic hydrograph separations were carried out using a constant composition groundwater and a variable composition rainfall at the Hsuechia experimental site during rainfall periods and the summer storms. Soil water, groundwater and rainwater samples were collected and preserved in polyethylene tubes in the dark at 4 °C for analysis. Oxygen isotopic variations were measured using the Epstein–Mayeda technique (Epstein & Mayeda 1953) and hydrogen isotopic compositions were determined after the reduction of water to H2 using zinc shots provided by the Biogeochemical Laboratory of Indiana University, USA (Coleman et al. 1982). Both oxygen and hydrogen analyses were conducted with isotopic ratio mass spectrometers at the Isotope Hydrology Laboratory of Academia Sinica, Taipei, Taiwan. The detailed methodology is described in the work of Sengupta et al. (2014). All δD and δ18O values are expressed relative to VSMOW in ‰ (per mil):
formula
where the R values represent the ratio of deuterium to hydrogen or 18O to 16O for the sample. All isotopic ratios are represented as the δ-notation (‰) relative to the international VSMOW (Vienna Standard Mean Ocean Water) standard and normalized on the scale that the δ18O and δD of SLAP (Standard Light Antarctic Precipitation) are −55.5‰ and −428‰, respectively (Gonfiantini 1978). The analytical precisions expressed as 1σ for the laboratory standards are ±1.3 ‰ for δD and ±0.08 ‰ for δ18O.
All rainwaters are depleted in their oxygen-18 and deuterium isotopes relative to the standard, resulting in negative δ values. The concentrations of oxygen-18 and deuterium in rainwaters are directly proportional to the ambient air temperature at the time precipitation is formed (Maule et al. 1994). However, they may vary with different positions and isotopic concentration of the atmospheric water vapor (Gat 1980). The Global Meteoric Water Line (GMWL) indicates specific atmospheric conditions at the source region of precipitation and/or at the site of its collection (Gat & Gonfiantini 1981; Ingraham 1998). The regression line derived from the 20-year δD and δ18O precipitation data all around the world is:
formula
The slope and intercept of Taiwan local meteoric water line (LMWL) slightly deviated from the GMWL. The local LMWL in Taiwan also revealed discrepancies with previous studies (Peng et al. 2010, 2012). The relationship between δ18O and δD in rainwater of Taiwan can be described by the linear regression with the equation:
formula

Isotopic compositions in precipitation

A total of 20 rainfall samples at the Hsuechia experimental site were analyzed for isotopic compositions (Figure 2) during rice growing season. The temporal variations of hydrogen and oxygen isotopic compositions of precipitation from July 2014 to September 2014 revealed that the δD of precipitation ranged from −21 ‰ to −82 ‰ with a mean of −41 ‰ ± −20 ‰ and the δ18O of precipitation ranged between −2.3‰ and −11.8 ‰, with a mean of −6.3 ‰ ± −2.5 ‰ (Figure 2), compared to the δD of global precipitation ranging from 0‰ to −64 ‰ with a mean of −25 ‰, and oxygen isotope compositions from −1.3 ‰ to −8.9 ‰ with a mean of −4.5 ‰ as reported by the International Atomic Energy Agency. This is supported by Wang & Peng (2001) who revealed that the large amplitudes of isotopic variations clearly indicate the complexity of its source and climatic characteristics in the precipitation of southwestern Taiwan. For example, the δD in the dry season (October–April) in Taipei of northern Taiwan ranged between 0 ‰ and −26 ‰ with a mean of −14 ‰ ± −8 ‰, whereas the δ18O was between −1.2 ‰ and −4.9 ‰ with a mean −3.3 ‰ ± −0.9 ‰ (Wang & Peng 2001). For the precipitation in Hualian of eastern Taiwan, on the other hand, the δD in the dry season ranged between −70‰ and −23 ‰ with a mean of −4 ± −16 ‰, whereas the δ18O ranged between −9.6 ‰ and −1.5 ‰ with a mean of −2.3 ‰ ± −1.9 ‰. The oxygen and hydrogen isotope compositions in the Hsuechia experimental site were relatively depleted in the rainfall. This result implies that elevated temperatures led to enriched stable isotopic composition in precipitation. Further, in the wet season oxygen and hydrogen isotope compositions in precipitation were significantly depleted, indicating that oxygen and hydrogen isotope compositions were more depleted in summer (wet seasons) than in winter (dry season). In addition, the significantly depleted isotopic composition occurs in heavy rainfall events over a relatively short time (Dansgaard 1964). Rainfall and temperature may have the mutual effects on isotopic composition, in which temperature has greater influence than rainfall does on a large scale. For all storms, groundwater recharge increased shortly after the onset of the storm, whereas the infiltration depth observed in soil profiles was only a few centimeters for small storms.

Figure 2

LMWL represents the local meteoric water line, dashed line based on rainwater data from 1991 to 2005, δD = 8.03δ18O + 11.11, n = 396, R2 = 0.98.

Figure 2

LMWL represents the local meteoric water line, dashed line based on rainwater data from 1991 to 2005, δD = 8.03δ18O + 11.11, n = 396, R2 = 0.98.

Close modal

Isotopic compositions in soil water

A total of 41 soil water samples at the Hsuechia experimental site were analyzed for isotopic compositions (Figure 3). The residence time of soil water carries very important and practical implications for tracing the source of water. The δD of the soil water was between −53‰ and −21‰ with a mean of −35 ‰ ± −7 ‰, whereas the δ18O ranged between −8.3 ‰ and −2.2 ‰ with a mean −4.7 ‰ ± −1.2 ‰. Figure 4 displays the depth profiles of hydrogen and oxygen isotopic compositions in the soil water of the studied site after the first storm (Typhoon Matmo). The distribution of δD and δ18O values with depth in soil water was significantly changed with rainwater. The depth-wise δD values ranged from −24 to −31 ‰, whereas the δ18O value ranged from −2.3 to −3.9 ‰ after the first storm. The highest water content in top soil was in the range of 97.2–99.3%. The δD and δ18O in soil water of the top layer were close to those in rainwater after the storm. The δD in soil water in the deepest soil layer (60–90 cm) approached a heavier value, from −28 to −31 ‰, differing from those of the rainwater. The results revealed that deep soil water was still occupied by the pre-event water, which is supported by Zhao et al. (2013) who found that deep soil water was replenished by preferential flow that bypassed the soil matrix at depth of 10–30 cm. A similar situation occurred in the δ18O profile.

Figure 3

(a) The isotopic compositions of precipitation with time at the Hsuechia experimental site, (b) isotopic compositions of deuterium and oxygen-18 in precipitation. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Figure 3

(a) The isotopic compositions of precipitation with time at the Hsuechia experimental site, (b) isotopic compositions of deuterium and oxygen-18 in precipitation. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Close modal
Figure 4

Depth-wise δD (a) and δ18O (b) distribution in soil water along the soil profile in the studied site after the July 23 storm (Typhoon Matmo). The soil water samples were collected on 30 July 2014.

Figure 4

Depth-wise δD (a) and δ18O (b) distribution in soil water along the soil profile in the studied site after the July 23 storm (Typhoon Matmo). The soil water samples were collected on 30 July 2014.

Close modal

Rainwater samples were collected and analyzed between July and October in 2014. Several sporadic rainfall events occurred between the first and second storms. The variation of groundwater level with rainfall during the experimental period (23 July–16 Sept 2014) (Figure 5(a)) demonstrated that the highest groundwater level occurred on 11 Aug and 12 Aug in 2014 after heavy rainfall. Figure 6 displays the δD and δ18O depth profiles of the soil water before and after the second storm (Typhoon Fung-Wong). The isotopic values varied significantly with depth. Before the second storm occurred, the isotopic compositions of soil water showed very little variation along the soil profile, the δ18O and δD values in the top soil layer (0–30 cm) were also high. On the contrary, the depth-wise δ18O and δD distributions of soil water were significantly changed after the storm event. The isotopic compositions of soil water in the top soil layer approached the same as in rainwater after the second storm due to rainwater infiltration directly. The δD in soil water at depth of 20–40 cm showed relatively lighter values, from −80 to −50‰, and moving away from those in rainwater. The δD values of deep soil water (60–80 cm) were close to those values before the storm (−36‰). This finding suggests that the water in the deeper soil layer was primarily occupied by the pre-event water (old water) after the storm. A similar pattern can be observed in the δ18O profile. After the second storm, the δD along the soil profile showed a increasing trend (Figure 7(a)), especially at 40–90 cm depth. The mean δD value of soil water was −52 (±− 15) ‰, suggesting it is affected by the groundwater. Further, soil water samples were collected four times repeatedly between the first and second storms in August 2014, due to the wet season. The δ18O values in soil water were enriched in the shallow soil layer at the upper 20 cm depth (Figure 7(b)). Similar results were reported by Gazis & Feng (2004), who found that the net contribution of evaporation is an enrichment of heavy isotopes near the soil surface. However, the reverse situation occurred in the profiles between 15 and 29 October (Figure 8). While there was no rainfall for almost one month, the paddy field was mainly irrigated with groundwater during this period. This suggests that after irrigation, groundwater played an important factor in mixing and controlling the isotopic composition of soil water in the studied site. The top soil at depth of 0–20 cm which is above the plough pan was probably having different water sources mixed in pores. The results of the present study suggest that the δD and δ18O values were mainly influenced by evaporation in the top soil layer and infiltrating water was significantly diminished as it reached 40–50 cm depth.

Figure 5

(a) Precipitation distribution at the studied site between July 2014 and October 2014; (b) Average monthly temperature (°C) and evaporation (mm) at the studied site between July 2014 and October 2014.

Figure 5

(a) Precipitation distribution at the studied site between July 2014 and October 2014; (b) Average monthly temperature (°C) and evaporation (mm) at the studied site between July 2014 and October 2014.

Close modal
Figure 6

Depth-wise δD (a) and δ18O (b) distribution in soil water along the soil profile at the studied site before and after 20 September 2014 (Typhoon Fung-Wong). The soil water samples were collected on 12 September and 23 September in 2014.

Figure 6

Depth-wise δD (a) and δ18O (b) distribution in soil water along the soil profile at the studied site before and after 20 September 2014 (Typhoon Fung-Wong). The soil water samples were collected on 12 September and 23 September in 2014.

Close modal
Figure 7

Soil water δD (a) and δ18O (b) variations with depth at additional four sampling dates after the first storm between 14 August and 18 September 2014.

Figure 7

Soil water δD (a) and δ18O (b) variations with depth at additional four sampling dates after the first storm between 14 August and 18 September 2014.

Close modal
Figure 8

Soil water δD (a) and δ18O (b) compositions with depth at the studied site during 15–29 October 2014 after the rainfall event (Typhoon Fung-Wong) on 20 September 2014.

Figure 8

Soil water δD (a) and δ18O (b) compositions with depth at the studied site during 15–29 October 2014 after the rainfall event (Typhoon Fung-Wong) on 20 September 2014.

Close modal

Figure 9 shows that the soil water in the top layer suffered from strong evaporation, shifting away from the LMWL. The isotopic compositions in the top soil layer were affected by strong evaporation after the first storm. The isotopic compositions of soil water in the deeper layers remained close to the LMWL, implying that the soil water at those depths was influenced very little by evaporation. Komor & Emerson (1994) also indicated that an evaporation front exists in a sub-horizontal surface, above which water is transported mainly as vapor, and below the front water is transported mainly as liquid. The plough pan was dominated by irrigation water and recent precipitation, indicating the existence of preferential flow in this layer. In this study, the isotopic compositions of soil water in the deep soil layer were relatively heavier than those in the shallow layer. The top soil layer has more structured macropores that are filled with mobile soil water (Zhao et al. 2013). Furthermore, Gazis & Feng (2004) found that the soil water at 60 cm depth in fact consisted of rainfall in an earlier period, implying the residence time of at least 3 months for that soil water. In addition, similar results were reported by Tang & Feng (2001) that the residence time of rainwater in deep soil water lasted for 4.5 months. Our results in this study showed that residence of earlier rainfall was present in most soil layers.

Figure 9

Isotopic compositions of deuterium and oxygen-18 in soil water. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Figure 9

Isotopic compositions of deuterium and oxygen-18 in soil water. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Close modal

This study revealed that the enrichment of isotopic compositions in the upper soil layer were caused by several parameters, such as evaporation, change in isotopic composition of different water sources, and mixing of new and old rainwater. Apart from evaporation, varied isotopic compositions in rainfall, irrigation input, and mixing of different sources of water are also important parameters that influence the isotopic composition of soil water in paddy fields. The evaporation significantly impacted the isotopic compositions of shallow soil water. Gazis & Feng (2004) found that light rainfall (less than 44 mm rainfall) was not displaced by preexisting immobile water in the upper soil layer. Significant variations in the soil profile indicate the existence prior to the current event.

Table 2

Environmental conditions of the experimental site during the observation period in 2014a

MonthTemperature (°C)
Relative humidity (%)
Average evaporation (mm)bSunshine (h d−1)Solar radiation (cal cm−2 d−1)
Max.Min.9:00 AM2:00 PM
July 34.6 24.7 93.7 78.3 6.2 4.3 167.4 
August 34.4 24.1 100 93.7 4.7 4.2 138.4 
September 34.4 23.1 88.9 85.2 4.7 4.8 129.0 
October 33.0 18.3 82.3 74.9 5.5 8.1 140.3 
MonthTemperature (°C)
Relative humidity (%)
Average evaporation (mm)bSunshine (h d−1)Solar radiation (cal cm−2 d−1)
Max.Min.9:00 AM2:00 PM
July 34.6 24.7 93.7 78.3 6.2 4.3 167.4 
August 34.4 24.1 100 93.7 4.7 4.2 138.4 
September 34.4 23.1 88.9 85.2 4.7 4.8 129.0 
October 33.0 18.3 82.3 74.9 5.5 8.1 140.3 

aSource: Workstation of Hsuechia. Chianan Irrigation Association of Taiwan.

bDiameter of evaporation dish is 120 mm.

Table 3

The duration and maximum intensity of rain events

Event dateRainfall (mm)Duration (h)Maximum rain intensity (mm/h)
23 Jul. 2014 38 16.5 15.6 
11 Aug. 2014 386.5 48 86.5 
20 Sep. 2014 36 14 12.5 
Event dateRainfall (mm)Duration (h)Maximum rain intensity (mm/h)
23 Jul. 2014 38 16.5 15.6 
11 Aug. 2014 386.5 48 86.5 
20 Sep. 2014 36 14 12.5 

Isotopic compositions in groundwater

The δD values of the groundwater at the Hsuechia experimental site of Chianan plain were between −43 ‰ and −42 ‰ with a mean of −43 ± −0.5 ‰. The δ18O values ranged between −6.3 ‰ and −5.9 ‰ with a mean of −6.1 ± −0.1 ‰ (Figure 10). The isotopic composition of hydrogen and oxygen in groundwater lies close to the local meteoric water line (LMWL) (Figure 10(b)), indicating that groundwater was derived mainly from rainwater. The isotopic compositions of soil water depend on mixing rates of surface runoff, interflow and groundwater in the soil layer. The difference in isotopic compositions changed with intensity of rainfall, residence time of interflow and intensity of evaporation in the soil layer. The isotopic compositions of hydrogen and oxygen revealed that groundwater in the studied site was mainly derived from the meteoric input. The main source of groundwater depends mainly on river water rather than rainwater recharge. There was a light rainfall event (73.9 mm) at the studied site in July 2014, in contrast to heavy rainfall in Aug 2014. The results indicated that two-month rainfall for groundwater recharge was limited, suggesting that groundwater in the studied site was recharged from earlier rains and/or river water from Jiangjiun River (Figure 1(a)).

Figure 10

(a) Temporal δD and δ18O isotopic compositions in groundwater at the studied site, (b) isotopic compositions of deuterium and oxygen-18 of groundwater. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Figure 10

(a) Temporal δD and δ18O isotopic compositions in groundwater at the studied site, (b) isotopic compositions of deuterium and oxygen-18 of groundwater. GWML represents the global meteoric water line (dashed line, δD = 8.17δ18O + 10.56) and LMWL is the local meteoric water line (solid line, δD = 7.9δ18O + 17.1).

Close modal

Seasonal and spatial variations

The isotopic characteristics of rainwater, soil water and groundwater in the experimental field, including their weighted mean values of δ18O, δD, and deuterium excess (d-excess) are summarized in Table 4. The d-excess of a sample is defined as d=δD − 8 × δ18O (Dansgaard 1964). Data of experimental period are not included for seasonal comparison to avoid the overlap of seasonal signals. In addition, southwestern Taiwan in winter falls in the dry season that has too few rainfall events to obtain meaningful statistics. The temperature effect is the prominent factor in the stable isotope signal. The results reveal that gradients for δ18O vs. temperature were about 0.31/°C for summer rainfall (data not shown). This is supported by Peng et al. (2010), who found that the gradient for δ18O vs. temperature is about 0.38/°C in summer precipitations. Compared with the reported data of Peng et al. (2010), the results indicated that δ18O values are relatively more depleted than those of winter rainfall in Taiwan. By comparing the isotopic compositions of groundwater and storm (event water), the event water contributing to groundwater was attributed to direct rainfall. For all storms, groundwater recharge increased shortly after the onset of the storm; however, light rainfall or small storms contributed smaller amounts in infiltration depth observed in the soil profile and indicated that light rains (<5 mm day−1) are less likely to infiltrate deep into the soil profile, resulting in water retention on plant and soil surface and subject to subsequently evaporating.

Table 4

Different species of water their weighted mean δ18O, δD and d-excess and related isotopic characteristics

Water speciesWeight mean values
MWLa
Nb
δ18O(‰)δD(‰)dc-excess(‰)SI
Rainwater −6.4 ± 2.5 −41.1 ± 19.9 10.1 7.8 8.6 19 
Soil water −4.7 ± 1.3 −35.3 ± 7.3 2.3 5.8 41 
Groundwater −6.1 ± 0.1 −42.5 ± 0.5 6.3 1.6 32.6 
Water speciesWeight mean values
MWLa
Nb
δ18O(‰)δD(‰)dc-excess(‰)SI
Rainwater −6.4 ± 2.5 −41.1 ± 19.9 10.1 7.8 8.6 19 
Soil water −4.7 ± 1.3 −35.3 ± 7.3 2.3 5.8 41 
Groundwater −6.1 ± 0.1 −42.5 ± 0.5 6.3 1.6 32.6 

aMWL, meteoric water line. S, slope; I, intercept.

bN, number of sample.

cd = δD − 8δ18O.

Stable isotopic composition of water can be used to reveal information about several hydrological processes in soil, including evaporation, infiltration, and groundwater flow. This study examined the hydrogen and oxygen isotopic compositions of rainwater, soil water, and groundwater in Chianan plain indicating that rice grown in the studied site didn't cause marked isotopic variations. Fertilization did not have a significant impact on isotopic compositions of soil water either. Except for irrigation water, the rainwater is another source for the generation of subsurface flow. Different sources of water during two storms were evaluated for their isotopic compositions in the studied site. The observed soil water processes had significant effect on groundwater recharge.

Evaporation primarily caused the enrichment of isotopic composition in the shallow soil layer. The δD and δ18O values of precipitation and groundwater match well in this study. The results revealed that intermittent light rain merely moistened top soil and then evaporated. Excessive rainfall intensity as exceeding infiltration capacity may generate surface runoff and is unfavorable for groundwater recharge. Consecutive rainfall is the most favorable feature for groundwater recharge.

Rainfall is one of the main sources of groundwater recharge, and affects the development of underground resources. The development and application of water resources, especially in countries or regions where water is scarce, should be more cautious and use detailed estimates. This research describes the relationship between rainfall and groundwater recharge, revealing the correlation between runoff loss and groundwater recharge. According to the amount of rainfall in the whole year the amount of groundwater recharge can be assessed. The amount of groundwater available must be clearly estimated to avoid serious consequences of over-pumping groundwater, for example land subsidence.

The authors are grateful to the Ministry of Science and Technology for providing funds to support this study. We thank the Chianan Irrigation Association of Taiwan for providing paddy fields to carry out the experiments. We thank Hui-Long Yang and Yi-Ting Chiu from the Chianan Irrigation Association of Taiwan for their technical assistance in sampling.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Bengtsson
L.
,
Saxena
R. K.
&
Dressie
Z.
1987
Soil-water movement estimated from isotope tracers
.
Hydrological Sciences Journal-Journal Des Sciences Hydrologiques
32
,
497
520
.
doi: 10.1080/02626668709491208
.
Chou
M.-L.
,
Jean
J.-S.
,
Sun
G.-X.
,
Hseu
Z.-Y.
,
Yang
C.-M.
,
Das
S.
&
Teng
J.-H.
2014
Distribution and accumulation of arsenic in rice plants grown in arsenic-rich agricultural soil
. Agronomy Journal 106(3), 945–951.
Coleman
M. L.
,
Shepherd
T. J.
,
Durham
J. J.
,
Rouse
J. E.
&
Moore
G. R.
1982
Reduction of water with zinc for hydrogen isotope analysis
.
Analytical Chemistry
54
,
993
995
.
doi: 10.1021/ac00243a035
.
Dansgaard
W.
1964
Stable isotopes in precipitation
.
Tellus
16
,
436
468
.
Dawson
T. E.
&
Ehleringer
J. R.
1991
Streamside trees that do not use stream water
.
Nature
350
,
335
337
.
Dincer
T.
,
Al-Mugrin
A.
&
Zimmermann
U.
1974
Study of the infiltration and recharge through the sand dunes in arid zones with special reference to the stable isotopes and thermonuclear tritium
.
Journal of Hydrology
23
,
79
109
.
http://dx.doi.org/10.1016/0022-1694(74)90025-0
.
Epstein
S.
&
Mayeda
T.
1953
Variation of O18 content of waters from natural sources
.
Geochimica et Cosmochimica Acta
4
,
213
224
.
http://dx.doi.org/10.1016/0016-7037(53)90051-9
.
Gat
J. R.
1980
The isotopes of hydrogen and oxygen in precipitation
. In: P. Fritz & J.-C. Fontes (Eds) Handbook of Environmental Isotope Geochemistry, vol. 1. The Terrestrial Environment.
Gat
J. R.
&
Carmi
I.
1970
Evolution of the isotopic composition of atmospheric waters in the Mediterranean Sea area
.
Journal of Geophysical Research
75
(
15
),
3039
3048
.
Gat
J. R.
&
Gonfiantini
R.
1981
Stable Isotope Hydrology: Deuterium and Oxygen in the Water Cycle. International Atomic Energy Agency (IAEA)
.
Gat
J.
&
Tzur
Y.
1967
Modification of the isotopic composition of rainwater by processes which occur before groundwater recharge
. Proc. Symp. Isotopes in Hydrology, Vienna, 1966, International Atomic Energy Agency, Vienna, pp. 49–59.
Gvirtzman
H.
,
Ronen
D.
&
Magaritz
M.
1986
Anion exclusion during transport through the unsaturated zone
.
Journal of Hydrology
87
,
267
283
.
Hooper
R. P.
,
Christophersen
N.
&
Peters
N. E.
1990
Modelling streamwater chemistry as a mixture of soilwater end-members – an application to the Panola Mountain catchment, Georgia, U.S.A
.
Journal of Hydrology
116
,
321
343
.
http://dx.doi.org/10.1016/0022-1694(90)90131-G
.
Ingraham
N. L.
1998
Isotopic variations in precipitation
. In: Isotope Tracers in Catchment Hydrology (pp. 87–118). Elsevier.
Kendall
C.
,
McDonnell
J. J.
&
Gu
W.
2001
A look inside 'black box' hydrograph separation models: a study at the Hydrohill catchment
.
Hydrological Processes
15
(
10
),
1877
1902
.
Laudon
H.
&
Slaymaker
O.
1997
Hydrograph separation using stable isotopes, silica and electrical conductivity: an alpine example
.
Journal of Hydrology
201
,
82
101
.
doi: 10.1016/s0022-1694(97)00030-9
.
Maule
C. P.
,
Chanasyk
D. S.
&
Muehlenbachs
K.
1994
Isotopic determination of snow-water contribution to soil-water and groundwater
.
Journal of Hydrology
155
,
73
91
.
doi: 10.1016/0022-1694(94)90159-7
.
Orlowski
N.
,
Lauer
F.
,
Kraft
P.
,
Frede
H. G.
&
Breuer
L.
2014
Linking spatial patterns of groundwater table dynamics and streamflow generation processes in a small developed catchment
.
Water
6
,
3085
3117
.
doi: 10.3390/w6103085
.
Peng
T.-R.
,
Wang
C.-H.
,
Huang
C.-C.
,
Fei
L.-Y.
,
Chen
C.-T. A.
&
Hwong
J.-L.
2010
Stable isotopic characteristic of Taiwan's precipitation: a case study of western Pacific monsoon region
.
Earth and Planetary Science Letters
289
,
357
366
.
doi: 10.1016/j.epsl.2009.11.024
.
Peng
T.-R.
,
Huang
C.-C.
,
Wang
C.-H.
,
Liu
T.-K.
,
Lu
W.-C.
&
Chen
K.-Y.
2012
Using oxygen, hydrogen, and tritium isotopes to assess pond water's contribution to groundwater and local precipitation in the pediment tableland areas of northwestern Taiwan
.
Journal of Hydrology
450–451
,
105
116
.
http://dx.doi.org/10.1016/j.jhydrol.2012.05.021
.
Sengupta
S.
,
Sracek
O.
,
Jean
J. S.
,
Lu
H. Y.
,
Wang
C. H.
,
Palcsu
L.
,
Liu
C. C.
,
Jen
C. H.
&
Bhattacharya
P.
2014
Spatial variation of groundwater arsenic distribution in the Chianan Plain, SW Taiwan: role of local hydrogeological factors and geothermal sources
.
Journal of Hydrology
518
(
Part C
),
393
409
.
http://dx.doi.org/10.1016/j.jhydrol.2014.03.067
.
Shanley
J. B.
,
Kendall
C.
,
Smith
T. E.
,
Wolock
D. M.
&
McDonnell
J. J.
2002
Controls on old and new water contributions to stream flow at some nested catchments in Vermont, USA
.
Hydrological Processes
16
,
589
609
.
doi: 10.1002/hyp.312
.
Tang
K.
&
Feng
X.
2001
The effect of soil hydrology on the oxygen and hydrogen isotopic compositions of plants’ source water
.
Earth and Planetary Science Letters
185
(
3–4
),
355
367
.
Vogel
T.
,
Sanda
M.
,
Dusek
J.
,
Dohnal
M.
&
Votrubova
J.
2010
Using oxygen-18 to study the role of preferential flow in the formation of hillslope runoff
.
Vadose Zone Journal
9
,
252
259
.
doi: 10.2136/vzj2009.0066
.
Wang
C. H.
&
Peng
T. R.
2001
Hydrogen and oxygen isotopic compositions of Taipei precipitation 1990 to 1998
.
Western Pacific Earth Sciences
1
(
4
),
429
442
.
Zhao
P.
,
Tang
X. Y.
,
Zhao
P.
,
Wang
C.
&
Tang
J. L.
2013
Identifying the water source for subsurface flow with deuterium and oxygen-18 isotopes of soil water collected from tension lysimeters and cores
.
Journal of Hydrology
503
,
1
10
.
doi: 10.1016/j.jhydrol.2013.08.033
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).