APP下载

Hydrochemical and isotope characteristics of spring water discharging from Qiushe Loess Section in Lingtai,northwestern China and their implication to groundwater recharge

2017-12-16SONGChaoHANGuilinWANGPanSHIYingchunHEZe

地下水科学与工程(英文版) 2017年4期

SONG Chao, HAN Gui-lin, WANG Pan, SHI Ying-chun, HE Ze

1The Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences, Shijiazhuang 050803, China.

2School of Water Resources and Environment, China University of Geosciences (Beijing), Beijing 100083, China.

Abstract: The loess plateau in northwestern China with an area of 640 000 km2, which has developed the loess deposits with a thickness up to 200 m in typical areas, is regarded as a huge carbon stock like the karst area in southwestern China, and plays an important role in regional(even global) carbon cycle. But the spring discharging from loess is poorly known compared with karst spring so far. The objective of this study is to ascertain the characteristics and origin of spring at Qiushe Village, Lingtai County, Gansu Province by hydro-chemical and isotopic methods. The results show that the springs including LGQ, HMQ, YYQ and CZQ are the depression spring and belong to the same shallow aquifer with the well water JZJ. There are not distinct seasonal/diurnal-scale variations on the hydro-chemical characteristics of the spring water (LGQ, HMQ, YYQ, CZQ) and groundwater (JZJ). The hydro-chemical type of groundwater is Ca·Mg-HCO3. The D and O isotope ratios indicate that the precipitation is the main recharge source of groundwater in study area. And the results of tritium (TU) and Cl concentration suggest that the recharge cycle of groundwater may be more than 60 yrs. Our study shows that the water cycle in loess plateau including rainfall, infiltration, recharge and discharge exerts a continuous impact on carbon stock in loess, which should be paid more attention to in future research on the quantitative reconstruction of paleoclimate.

Keywords: Spring discharging from loess; Origin; Precipitation; Hydrochemistry; Shallow groundwater

Introduction

The China’s Loess Plateau (CLP) with an area of 640 000 km2has developed the loess deposits with a thickness up to 200 m in typical areas (LIU T, 1985). The content of CaCO3accounts for 12%-16.1% in loess layer and 49%-62% in CaCO3illuvial horizon (ZHAO J B, 2000), 79.2%-98%(89.7% on average) of which are secondary carbonate (WEN Q, 1989). Moreover, the data about hydro-chemical composition of surface water in China’s loess plateau show that the dominant anion is HCO3-. These facts reveal that the dissolution/precipitation of carbonate mineral is a key geochemical process in loess, and that the carbon stock in loess may play an important role in regional/global carbon cycle. The absorbed C in the loess area is estimated to be about l.76×1013gC(0.0176 PgC) annually if the whole loess area is covered by grassland, suggesting that the loess as a sink of atmospheric CO2should not be ignored in global C (QIN X et al. 2001).

The hydro-chemical composition of rainfallrecharging spring in karst area of southwestern China is usually used as an indicator to explore the kinetics of carbonate rock dissolution and the relative process of carbon cycle in karst area and to estimate CO2consumption by carbonate weathering (LIU Z H et al. 2010a, b; LI S L et al.2008; Sun H L and Liu Z H, 2010; ZHAO M et al.2015; ZHAO M et al. 2010), but few studies have been conducted incarbonate-rich loess plateau of northwestern China. In CLP, the origin and recharge source of groundwater in loess and springs discharging from loess valley are under debate.

The studies by LI Yu-shan et al. concluded that rainfall can hardly recharge groundwater of loess tableland by in-situ infiltration because of the prevention of the drying layer (LI Yu-shan, 1983;LI Y, 2001). WANG Z et al. (2000) found in a 2-year tritium tracing experiment that the rate of water infiltration was less than 0.05 m/yr and the infiltration capacity is less than 1% of rainfall. And the study of CHEN J et al. (2012) held that the local precipitation was not the main recharge source of groundwater in Ordos Plateau (on the north of the CLP), as the mean isotopic compositions of the local precipitation are significantly higher than those of groundwater.However, many recent studies have different findings. Numerous methods including the water balance (QU H, 1991), the tritium peak (LIN R and WEI K, 2006; ZHANG Z G et al. 1990) and the chlorine mass balance (Gates J B et al. 2011;HUANG T and PANG Z, 2011) were employed in different field sites of loess plateau to estimate the rate of groundwater recharge, and the rate of 33-94 mm/yr was obtained, which contributed to about 6.3-18% of local precipitation. The results from HUANG T et al. (2013) showed that the rate of groundwater recharge beneath natural uncultivated grassland is about 94-100 mm/yr.These differences are related to the amount of rainfall, soil type and thickness, as well as vegetation coverage.

As for the spring discharging from Loess profile, its characteristics and origins need to be explored firstly before the studies on carbon cycle are conducted. Therefore, the hydro-chemical analysis and isotope methods were used in this study to ascertain the characteristics and origins of springs discharging from a loess section in the center of CLP.

2 Methods

2.1 Study area

The study area is located in Qiushe (35°10′N,107°41′E), Lingtai County, Gansu Province, in northern China (Fig. 1). The land form element of Qiushe consists of a flat and long strip-shaped loess tableland with a width of 1.1-1.5 km and numerous gullies. The table land is about 1 225 m above sea level. Perennial spring is usually developed at the bottom of gully, as LGQ, HMQ,CZQ and YYQ in Fig. 1.

Lingtai Loess profile consists of the upper loess-paleosol sequence with a total thickness of 168 m and the lower 120-m-thick Red Clay formation (DING Z L et al. 1999; SUN D et al.1998; SUN Y et al. 2006). The Loess-Paleosol sequence, including the Wuchen Loess (L33-S15),the Lishi Loess (L15-S1), the Malan Loess (L1) and the Holocene Loess (S0), has accumulated from 2.6 Ma B.P. to the present (DING Z L et al. 1999).The soil of tableland is mainly used to plant wheat and maize.

At present, the East Asia monsoon influences the climate of the Lingtai area, generating a semi-humid, warm temperate and continental monsoon climate, which is characterized by relatively hot and humid summers and cold and dry winters. According to the meteorological data collected at Lingtai Station from 1959 to 2008, the mean temperature is 15.3 °C, with monthly mean annual temperature in July and January of 22.1 °C and -4.7 °C, respectively. Mean annual precipitation is 605.5 mm, with rainfall concentrated mainly in summer and autumn. Mean annual latent evaporation reaches 1 492 mm.

2.2 Sampling

The spring water and well water were collected in polyethylene bottles and filtered through 0.45 μm Millipore membrane for major element analyses. The spring water was sampled at the outlet. And the groundwater samples were taken from wells for domestic and agricultural purposes.The samples from the wells were taken by using pumps installed in these wells and the samples were taken after several well volumes were removed prior to sampling. Stable isotope samples were taken in 100-ml HDPE screw-cap bottles and sealed tightly to prevent evaporation. Samples for tritium analysis were collected in 500-ml polyethylene bottles.

Fig. 1 The location and hydrological sketch map of study area

2.3 Measurement

1) Hydro-chemical analysis

The temperature (T), pH, Dissolved Oxygen(DO), Electrical Conductivity (EC) and Oxidation-Reduction Potential (ORP) of spring and well water were measured by a digital multi-parameter meter (Multi 3430 WTW,Germany) at outlets and wellheads in field,respectively. Major chemical parameters like pH,K+, Ca2+, Na+, Mg2+, NH4+, NO3-, NO2-, free CO2,HCO3-, Cl-, SO42-, COD and HPO42-were analyzed.Here, Base cations (K+, Na+,Ca2+and Mg2+) were determined by inductively coupled plasma atomic emission spectrometry (ICP-AES) on a Thermo Scientific iCAP 6300 ICP spectrometer. NH4+was analyzed by means of Nessler’s reagent colorimetry. NO3-was determined with ultraviolet spectrophotometric method. NO2-was determined by α-naphthylamine colorimetric method. Free CO2, HCO3-and Cl-were tested with titration method. SO42-was determined by molybdenum blue spectrophotometric method. COD was determined by potassium permanganate method.HPO42-was determined by molybdenum blue spectrophotometric method. The error was always less than 3%.

2) δ2H and δ18O

Measurements of δ2H and δ18O were made on a MAT253 mass spectrometer with analytical precisions of better than 0.1% for oxygen and 2%for hydrogen, respectively. The results were reported in delta (d) notation asper mil (%)differences relative to standard values of VSMOW:

where R is the D/H ratio or the18O/16O ratio,respectively.

3) Tritium

Water samples for3H measurement were firstly distilled at atmospheric pressure with potassium permanganate as additive until their specific conductivity decreased to values below 5 ls/cm. In this procedure, the tritium content of the sample was unaltered with a nearly 100% recovery. Then,250 ml of the distilled water was electrolyzed.Finally, 8 ml of the electrolyzed water was mixed with 12 ml of scintillator cocktail in a 20-ml glass vial for tritium measurement. The tritium values were measured by a liquid scintillation analyzer(model: Quantulus 1220) at the Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences. Tritium concentrations are expressed as absolute concentrations, using tritium units (TU). The precision of measurements is ±1 TU.

3 Results and discussion

3.1 The seasonal, diurnal-scale hydrochemical variations of spring water: pH,T, DO, EC and ORP

The pH, T, DO, EC and ORP of spring, well water and rainfall are listed in Table 1. From Table 1, these indexes of spring water (LGQ,HMQ, CZQ and YYQ) do not show distinct differences on a time scale, exhibiting that the groundwater that feeds springs is of a relative stable environment. For QSJ, the EC and DO and ORP have evident differences between two periods,which is attributed to the different pumping environment conditions of wells. For the rainwater LTY has lower pH and EC, and higher DO than spring and well water. The difference of the pH, T,DO, EC and ORP among different springs can be observed because the biogeochemical conditions in soil may have different impacts at the outlet of spring.

The diurnal patterns of pH, temperature, DO and EC of spring water LGQ are plotted in Fig. 2.They show week differences at a diurnal scale. The water temperature, DO and EC pattern have a 0.1-0.2 degree, 2% and 2 μs/cm higher within the period of 10:00 am-5:00 pm than within the period of 5:00 pm-9:00 am while the pH pattern shows a 0.04 unit lower (Fig. 2). The weak (0.1-0.2 ºC)temperature difference of LGQ relates to the maximal absorption of solar radiation during the day, which may result in the slight variability of pH, EC and DO in LGQ.

These change characteristics of spring water in CLP is different from the results from the spring water in subtropical karst areas of southwestern China, where the pH, EC of spring water are very sensitive to atmospheric temperature, rainfall, soil CO2, and even the storm-events (LIU Z Het al.2007).

Table 1 The pH, temperature, DO, EC and ORP of spring and well water

Fig. 2 Diurnal change of T, pH, DO and EC in spring water

3.2 Major hydro-chemical characteristics and type

The results of major hydro-chemical parameters of spring water are listed in Table 2.The results show that the cations (K+, Na+, Ca2+,Mg2+) and the anions (Cl-, SO42-, HCO3-, NO3-) in different spring water during different seasons have a similar concentration. And NH4+, HPO42-and NO2-have low concentrations. The COD of LGQ, HMQ in summer is a little higher than in winter because of the mixture of alluvium deposit in rainy season. Free CO2and the pH also exhibit a similar characteristic in different spring water and during different seasons. For well water JZJ, the major hydro-chemical parameters are similar to the spring water. Whereas, rain water has a lower concentration of cations and anions, lower pH, and lower free COD and higher COD than spring water.

Water type in loess is dependent on the factors like mineral components of aquifers, retention time and flow pattern of groundwater flow. In order to identify the dominant water type of spring water and groundwater, all the samples were plotted in a piper diagram (see Fig. 3) (Piper A M, 1944). The concentration of cations and anions are plotted in piper. Showing relatively same results, all the water samples are categorized in Ca·Mg-HCO3. It shows that the hydro-chemistry of groundwater in this study is controlled by carbonate weathering.

3.3 Recharge source of shallow groundwater: Precipitation

The δ2H and δ18O as excellent tracers for determining the origin of groundwater are widely used to explore the natural water cycle and groundwater movement (CHEN Z et al. 2011;Craig H, 1953; Craig H, 1961; Simmons C T et al.2002). In this study, δ2H and δ18O values of all water samples have ranges of -72‰<δ2H<-70‰and -10.1‰<δ18O<-9.4‰, respectively (Table 3).The δ2H-δ18O plots of all spring water and groundwater are shown in Fig. 6, where the equation of LMWL is δ2H = 7.44 δ18O+1.69 (with R2=0.94, n=30) (WANG Ret al. 2008). The isotope compositions are close to the GMWL(Fig. 4), which is consistent with other results(HUANG Tet al. 2013) from the Ansai terrace(1 100-1 200 m; (Gates J Betal. 2011)), the Chabagou hilly region (898-1 302 m; (LIU Xet al.2011)); and typical loess plains,e.g. Changwu(-1 200 m; (WANG R, 2007)), Xifeng (-1 420 m),Luochuan (-1 150 m; (HUANG T, 2010)) and Xi’an (400-1 200 m; (QIN Da-jun and TAO Shu-hua, 2001)). In Fig. 4, all spots of spring water and groundwater form a cluster together on the LMWL, indicating a common origin for both groundwater and spring water, namely precipitation. This is supported by other study methods such as water balance (QU H, 1991),tritium peak (LIN R and WEI K, 2006; ZHANG Z Get al. 1990) and chlorine mass balance (Gates J Bet al. 2011; HUANG T and PANG Z, 2011).

3.4 Recharge time of groundwater

The tritium (3H) content of spring water and groundwater is below 4.4 TU (Table 3), indicating that the groundwater is largely recharged before nuclear tests (1950s). This is consistent with the tritium results in many loess plains (e.g. Xifeng,Luochuan, Fuping and Weinan) in the LPC(HUANG T, 2010; TIAN Hetal. 2007). Whereas the study of ZUO Ret al. (2011) pointed out the retardation effect of tritium migration in China’s Loess media because the distribution coefficient of tritium is influenced by initial activity of tritium,pH, water/solid ratio, and the content of humic and fulvic acids (ZUO Ret al. 2011). According to other studies in different field sites of loess plateau(ZHANG Z Get al. 1990; QU H, 1991; LIN R and WEI K, 2006; Gates J Bet al. 2011; HUANG T and PANG Z, 2011; HUANG Tet al. 2013), the groundwater recharge rate in Qiushe area (a typical loess area) may be about 100 mm/yr.

Table 2 Major hydro-chemical parameters of spring water (mg·l-1)

Fig. 3 Piper diagram of spring water and groundwater

Fig. 4 The δ2H-δ18O relationship of shallow groundwater from the CLP

In addition, the chloride mass balance (CMB)technique is now widely and successfully used to estimate groundwater recharge in (semi) arid areas(Scanlon B Ret al. 2006). YIN Li-heet al.(2010)used the CMB between precipitation and groundwater to estimate the recharge rate (5-74 mm·yr-1) in the Ordos Basin where precipitation ranges from 189 to 342 mm·yr-1. In this study,when compared with the CMB of soil profile in loess unsaturated zone (HUANG Tet al. 2016), the low Cl concentration of groundwater (ranging from 5.52 to 13.03 mg·l-1) also implies that modern water (since the 1950s) has not entered aquifer yet.

4 Conclusions

The hydrochemical characteristics of the springs water (LGQ, HMQ, YYQ, CZQ) and groundwater (JZJ) fail to display distinct seasonal/diurnal-scale variations. The hydro-chemical type of groundwater is Ca·Mg-HCO3, and the main geochemical process is controlled by carbonate weathering.

All spots of D and O isotope data of spring water and well water form a cluster together on the LMWL, indicating that the precipitation is the main recharge source of groundwater in study area.Tritium (TU) free in spring water and well water suggest that the recharge cycle of groundwater may be more than 60 yrs, which is attributed to the long-term groundwater recharge as a result of a thick loess unsaturated zone with many paleosol layers. This is consistent with the low Cl concentration and stable hydro-chemical characteristics in groundwater.

Table 3 δ2H and δ18O and tritium

Like the springs in karst area of Southwestern China, the springs as a medium for excavating the biogeochemical information in loess deposit, can be also used to study the weathering of loess carbonate and the consumption of atmospheric/soil CO2by carbonate weathering. Moreover, CaCO3content in the loess is regarded as a proxy index of precipitation: A higher rainfall will cause stronger leaching of soluble calcium carbonate. On the other hand, a higher content of calcium carbonate is associated with the lower annual precipitation.However, the continuous impact of water cycle including rainfall, infiltration, recharge, discharge on CaCO3should not be ignored in future research about the quantitative reconstruction of paleoclimate.

Acknowledgements

This work was supported by the Chinese National Natural Science Foundation (No.414 03107 and No.41325010), and the Basic Science Research Fund from the Institute of Hydrogeology and Environmental Geology (No. SK201208) and China Geological Survey Projects (No. 1212011 3005900).


登录APP查看全文