COBISS: 1.01

The effects of agricultural activities and atmospheric acid deposition on carbonate weathering in a small karstic agricultural catchment, Southwest China 
Vpliv kmetijske dejavnosti in kislih usedlin iz zraka na preperevanje karbonatov na majhnem kraškem kmetijskem območju na jugozahodu Kitajske 
Yu Chen1 & Yongjun Jiang 1, 2*

Abstract	UDC  552.54:551.3.053(513)
Yu Chen & Yongjun Jiang: The effects of agricultural activities and atmospheric acid deposition on carbonate weathering in a small karstic agricultural catchment, Southwest China
In order to quantify the sources and fluxes of DIC, the effects of the use of N-fertilizers and acid deposition on carbon­ate weathering have been quantified by hydrochemistry and .13CDIC of groundwater in Qingmuguan underground river system (QURS) – a small karstic agricultural catchment of Southwest China. The results show that: (1) the significant tem­poral variations for major element concentrations and .13CDIC of groundwater in different months were observed, especially, of which the groundwater showed significant high concentra­tions of DIC, Ca2+, Mg2+, NO3-, SO42- and .13CDIC in rainy season and fertilizing period in the QURS; (2) the contributions of carbonate dissolution by carbonic acid to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months averaged 68.5 % and 81.0 %, respectively. While the contribu­tions of carbonate dissolution by nitric acid originated from the use of N-fertilizers and atmospheric acid deposition to to­tal concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months averaged 11.1 % and 6.7 %, respectively, and the contributions of carbonate dissolution by sulphuric acid originated from the atmospheric acid deposition to to­tal concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months averaged 20.4 % and 12.3 %, respectively; (3) the .13CDIC increased obviously with (Ca2++Mg2+)/HCO3- of groundwater in the rainy season and fertilizing period indicat­ed that the use of N-fertilizers and atmospheric acid deposition should be responsible for the elevated the .13CDIC and the molar ratio of (Ca2++Mg2+)/HCO3- of groundwater in the QURS. 
Key words: carbonate weathering, karst groundwater, agri­cultural activities, atmospheric acid deposition, Qingmuguan, Southwest China.
Izvleček 	UDK  552.54:551.3.053(513)
Yu Chen & Yongjun Jiang Vpliv kmetijske dejavnosti in kis­lih usedlin iz zraka na preperevanje karbonatov na majhnem kraškem kmetijskem območju na jugozahodu Kitajske
Da bi določili izvor in tok DIC (raztopljen anorganski ogljik) smo vrednotili učinek uporabe dušikovih gnojil in kislih usedlin na preperevanje karbonatov. Pri tem smo se poslužili hidrokemičnih postopkov in analize .13CDIC podzemne vode v kraškem zaledju reke Quinmugan (QURS), ki je del manjšega kmetijskega območja na jugozahodu Kitajske. Rezultati so pokazali naslednje: (1) Izmerjene so bile opazne časovne spremembe v koncentraciji glavnih elementov in .13CDIC v podzemni vodi v različnih mesecih; to je še posebej očitno v deževnem obdobju in času gnojenja, ko so bile zabeležene iz­razito povečane koncentracije DIC, Ca2+, Mg2+, NO3-, SO42- in .13CDIC; (2) Delež preperevanja karbonatov zaradi ogljikove kisline je glede na skupne koncentracije (Ca2++Mg2+) in HCO3- v podzemni vodi v različnih mesecih v povprečju znašal med 68,5 % in 81,0 %. Medtem je delež preperevanja karbonatov zaradi dušikove kisline (posledica uporabe dušikovih gnojil in kislih usedlin iz zraka) glede na skupne koncentracije (Ca2++Mg2+) in HCO3- v podzemni vodi v različnih mesecih v povprečju znašal med 11,1 % in 6,7 %, delež preperevanja karbonatov zaradi žveplove kisline (posledica kislih usedlin iz zraka) glede na skupne koncentracije (Ca2++Mg2+) in HCO3- v podzemni vodi v različnih mesecih pa je v povprečju znašal med 20,4 % in 12,3 %; (3) Vsebnost .13CDIC v podzemni vodi se v času deževne sezone občutno poveča skupaj z razmerjem (Ca2++Mg2+)/HCO3-, obdobje gnojenja pa nakazuje, da upo­raba dušikovih gnojil in usedanje kislin iz ozračja vplivajo na povišano vsebnost .13CDIC in molarno razmerje (Ca2++Mg2+)/HCO3- v podzemni vodi v rečnem sistemu QURS. 
Ključne besede: preperevanje karbonatov, kraška podzemna voda, kmetijske dejavnosti, kisle usedline iz zraka, Qingmu­guan, jugozahodna Kitajska.

1 School of Geographical Sciences, Southwest University, Chongqing 400715, China;
2 Karst Environment Laboratory, Southwest University, Chongqing 400715, China.
Yu Chen, School of Geographical Sciences, Southwest University, 400715 Chongqing, China; e-mail: chenyu192781@163.com
*Corresponding author: 
Yongjun Jiang, School of Geographical Sciences, Southwest University, 400715 Chongqing, China; e-mail: jiangjyj@swu.edu.cn
Received/Prejeto: 18.03.2016

ACTA CARSOLOGICA 45/1, 161–172, POSTOJNA 2016

Yu Chen & Yongjun Jiang

Introduction

The geological-environmental characteristics of karst hy­drological systems result in a highly vulnerable system, which is considerably sensitive to external environmental changes. While the increasing environmental pollution, both deliberate and unintentional forms as consequence of human activities, has to a great extent spoiled sensitive karst hydrological systems in Southwest China (Liu et al. 2006, 2008; Jiang et al. 2008, 2009a, 2009b; Jiang 2012; Pu et al. 2011). Concentrations of nitrate and sulphate of karst groundwater in southwest China increase notably as a result of large amount of chemical fertilizers used in agriculture (Liu et al. 2006; Jiang et al. 2008, 2009a, 2009b; Jiang 2012; Pu et al. 2011) and acid deposition (Li et al. 2008, 2010), which could not only influence the quality of karst groundwater, as well as the carbonate weathering process related to the global carbon cycle. 
Conventionally, the DIC in karst groundwater is dominantly derived from carbonate dissolution by car­bonic acid (Eq. 1), which forms from reaction of soil or atmospheric CO2 with water.
(Ca1-xMgx)CO3 + H2O + CO2› (1-x)Ca2+ + 	
xMg2+ + 2HCO3-	(1)
In this case, half of the DIC (generally, bicarbon­ate is the dominant DIC species in karst groundwater) in karst groundwater derived from soil/atmospheric CO2 constitutes an important sink of atmospheric CO2.
However, recently increases in the inorganic carbon flux in karst groundwater have been linked to agricul­tural activities and acid precipitation (Semhi et al. 2000; Calmels et al. 2007; Li et al. 2008, 2010; Perrin et al. 2008; Barnes & Raymond 2009; Ali & Atekwana 2011; Gandois et al. 2011; Jiang 2013; Yue et al. 2015). 
The consequence of the use of nitrogen fertilizers (usually in the form of (CO(NH2)2), (NH4)2SO4, NH3 and (NH4)2PO4, is the release of protons in the soil during the nitrification process. N fertiliser oxidation produces two protons for every nitrified ammonium ion and then en­hances carbonate weathering. These processes can be ex­pressed as following equations:
NH4+ + 2O2 = NO3- + 2H++ H2O		(2)
CaCO3 + H+ = Ca2+ + HCO3-		(3)
NH4+ + 2O2 + 2(Ca1-xMgx)CO3 = NO3- +
2(1-x)Ca2+ + 2 xMg2+ + 2HCO3- + H2O	(4)
Acid deposition, including reactive N and sulfate, has dramatically increased worldwide in the past few de­cades, of which more than one third of the territory in China (Tang et al. 2010), especially in Southwest China, is suffering from acid deposition. Anthropogenic Nr is released to the atmosphere either as nitrogen oxides (NOx), mainly from combustion, or as ammonia (NH3), mainly from agriculture (Dentener et al. 2006; Paulot et al. 2013). NOx is oxidized in the atmosphere to nitric acid (HNO3) and ammonia can be transformed into NH4+ through chemical reactions and then can be nitrified and produce proton. Sulfate is released to the atmosphere as SOx, predominately from sulfur emissions (fossil fuel combustion) (Dentener et al. 2006), and SOx can be oxi­dized in the atmosphere to sulfuric acid (H2SO4). Thus, acid deposition can impact carbonate weathering. These processes can be expressed as following equations:
(Ca1-xMgx)CO3 + HNO3 › (1-x)Ca2++ xMg2+ +	
NO3- + HCO3-	(5)
2(Ca1-xMgx)CO3 + H2SO4 › 2(1-x)Ca2+ + 	
2xMg2++ SO42- + 2HCO3-		(6)
In these cases, carbonate dissolution by nitric and sulphuric acids are resulting in greater DIC export, which is derived from the carbonate rather than from the CO2 sequestration. Such enhanced carbonate weath­ering by anthropogenic acidity inputs could not only in­fluence the element fluxes to riverine systems, as well as the global carbon cycle.
Thus, based on these knowledge, this paper pre­sented the results of chemical analysis for the concentra­tions of major ions, DIC and .13C-DIC of groundwater, and discussed: (1) to quantify the sources and fluxes of DIC in groundwater, and (2) to evaluate the effects of N-fertilizers and acid deposition on carbonate weather­ing in a small karstic agricultural catchment of South­west China.

The effects of agricultural activities and atmospheric acid deposition on carbonate ...

Study area

The Qingmuguan Underground River System (QURS) is located at the west of Chongqing municipality, Southwest China (Fig. 1). The underground drainage area of the sys­tem is approximately 11.4 km2. The elevation of QURS is between 320~640 m above average sea level. The climate is primarily subtropical monsoonal with a mean annual precipitation of 1100 mm and a mean air temperature of 16.5 °C. The monsoonal climate results in a rainy season from April to October and a dry season from November to March.
Geology and hydrogeology
The geologic layers of the study area are shown in Fig. 1. The QURS lies in the middle part of the Wentangxia an­ticline, Chongqing, and the aquifers are mainly underlain by Lower and Middle Triassic strata (Fig. 1). Carbonate rocks (limestone) cover an area of 10 km2 or about 88 % of the total area. The strata of the anticlinal axis are carbonate rocks of the Lower Triassic Jialingjiang For­mation (T1j), with limestone being a major lithology, whereas anticlinal wings are carbonate rocks of the Mid­dle Triassic Leikoupo Formation (T2l) and sandstones with some coal seams of the Upper Triassic Xujiahe Formation (T3xj). Yellow-green calcareous clay rocks of the Lower Leikoupo (T2l) overlie the Jialingjiang Group (T1j). No sulfate evaporates (gypsum and anhydrite) are exposed in the study area. Due to the banded distribution of carbonate rocks and the presence of relatively imper­meable sandstone at two wings combined with a vertical patulous cranny which was well developed in the anticli­nal core, basic conditions exist for a formation of a karst trough valley and the development of a karst groundwa­ter system. There are many depressions, caves, dolines and sinkholes in the catchment. In the upper stream, the surface water in Ganjiacao depression, which is the biggest depression, recharges the QURS via Yankou sink­hole. The Qingmuguan Underground River developed in the core of the karst trough valley and flowed along anticlinal axis (Lower Triassic Jialingjiang formation) in a NE-SW direction with a total length of 7.8 km. In the downstream reach of the QURS, the Jiangjia spring dis­charges the groundwater of the QURS. High-resolution tracer tests in the catchment indicated a single conduit between Yankou sinkhole and Jiangjia spring (Fig. 1) (He et al. 2010), and the Jiangjia spring is the only outlet of the hydrological system. Flow rates of the spring have large variations around a year from 0.002 to 3.5 m3/s, with an average of 20.5 l/s. The spring water is used by individual households.
Land use pattern, vegetation, soil 
and agricultural practices
There are three land use categories (Fig. 1): forested land, paddy land and dry land. The percentages of land use were 64.9 for forested, 11.4 for paddy land and 23.7 for dry land, respectively. Paddy fields are mainly distributed in the Ganjiacao depression, and dry fields are scattered around the bottom of the valley. The crops are rice in paddy field, potato and vegetables in dry field.
Vegetation is predominantly subtropical evergreen broadleaved forests. Those plants and crops are C3 pho­tosynthetic type.
The dominant types of soils are mainly limestone soil and yellow soil derived from carbonate rocks and sandstones, respectively. Paddy soil are scattered in the depressions.
Farmers principally spread chemical fertilizers on soils in spring and summer seasons, mainly compound fertilizers of the N-P-K type and straight fertilizers such as urea (CO(NH2)2), NH4HCO3 and ammonium sul­phate ((NH4)2SO4). Typical amounts of chemical fertil­izers used in the catchment are 1200 kg ha-1 yr-1 for rice, 600 kg ha-1 yr-1 for vegetables. Because soil pH and car­bonate content are high, buffering with lime is not prac­ticed.


Fig. 1: Location, hydrogeological and land use map in the QURS.

Yu Chen & Yongjun Jiang

Samples and analyses

The rainfall was collected from the HOBO weather sta­tion (made by OnSET Ltd., USA) installed in the catch­ment, which the resolution of rainfall was 0.2 mm with a time internal of 15 min. The discharge, water tempera­ture, pH and specific conductivity of groundwater were monitored at 15 min intervals using a Greenspan CTDP 300 multi-channel data logger. 
Monthly samples of groundwater were collected for analysis of major hydrochemical components and iso­tope in laboratory in 2013. Rainwater was collected dur­ing 3 events in the catchment from May to July, 2013.
Temperature, pH, specific conductivity (SC, at 25 °C) and dissolved oxygen (DO) of groundwater and rainwater were measured in the field using a WTW mul­tiparameter probe (Multiline P3 PH/LF-SET) with reso­lutions of 0.1 °C, 0.1, 1 µS cm-1 and 0.1 mg L-1, respec­tively. NH4+ of groundwater and rainwater was measured in the field using a D8500 water quality multimeter (HACH, America), with resolution of 0.001mg/l. Ca2+ and HCO3- were determined by a test kit with a titra­tion pipette (Aquamerck) in the field with resolutions of 2 mg/l and 0.1 mmol/l, respectively.
Water samples of the spring and rain were collected by injection syringes and were immediately filtered into pre-rinsed plastic containers (1 L) with 0.45 µm filter membranes for ion analyses, one of which was acidified to pH<2 with HNO3 to preserve cation concentrations. To limit gas exchange, the plastic containers were cooled with ice. Samples for the stable carbon isotope of dis­solved inorganic carbon (.13CDIC) analysis were filtered into 10 mL tubes with 0.2 µm cellulose-acetate .lters. Soon after collection, these samples were preserved us­ing three drops of a saturated solution of HgCl2 to pre­vent microbial alteration. All samples were kept refriger­ated (below 4 °C) until analysis.
Three representative plant samples (including the root, stem and leaf), rice, bamboo and Cyclobalanop­sis, were collected in the catchment. Meanwhile, three soil CO2 samples were collected, using a capillary tube reaching in 0.4-0.8 m depth and an evacuated glass ves­sel with Kontes valves, under the corresponding plants. Also, two limestone samples were collected from Lower Trissic Jialingjiang formation and Middle Triassic Leik­oupo Formation, respectively. 
Concentrations of major cations and anions were measured with inductively coupled plasma optical emission spectrometry (ICP-OES) with a resolution of 0.01 mg/l and ion chromatography (IC) with a resolu­tion of 0.01 mg/l, respectively, at the Water Environmen­tal Laboratory of Southwest University. The precision of the IC and ICP analyses was within ±5 % for major ele­ments.
For .13CDIC, using the modified method of Ate­kwana and Krishnamurthy (1998) for stable carbon isotope analysis of dissolved inorganic carbon, a 10 mL water sample was injected by syringe into glass bottles that were pre-filled with 1 mL 85 % phosphoric acid and magnetic stir bars. The CO2 was extracted and purified after cryogenic removal of H2O using a liquid nitrogen–ethanol trap. Finally, the CO2 was transferred cryogeni­cally into a tube for isotope measurement. 
Plant samples were ultrasonically cleaned for 15 min in deionized water, and dried in an oven at 50 °C for 48 h. Then, the plant samples were ground into pow­der with diameters less than 150 µm to ensure homoge­neity, of which 1 to 2 mg was placed into stannum cups for carbon isotope analysis.
All analyses for .13C were done at the Geochem­istry Laboratory of Southwest University. The carbon isotopic compositions of soil organic matter and plants were determined using an elemental analyzer coupled to an isotope-ratio mass spectrometer (EA-IRMS), and the carbon isotopic compositions of DIC, soil CO2 and car­bonate were determined using Gas Bench-linked with Delta V Plus gas stable isotope-ratio mass spectrometer (Gasbench-IRMS). The results were expressed on a con­ventional permil scale with respect to the Vienna Pee Dee Belemnite (V-PDB) standard. The overall experi­mental accuracy for .13C during this study was ±0.2 ‰. 

The effects of agricultural activities and atmospheric acid deposition on carbonate ...

Results and discussion

The .13C of limestone, plants and soil organic matter
Isotopic compositions (.13C) of limestone, plants and soil CO2 are shown in Tab. 1. The .13C of limestone from Ji­naglingjiang and Leikoupo information was 0.2 ‰ and 0.3 ‰, respectively, consistent with typical values of ma­rine limestone. The .13C of plants varied from -26.6 ‰ to -28.6 ‰ with an average value of -27.9 ‰, consist­ent with the major vegetation type using the C3 carbon fixation pathway in the catchment. The .13C of soil CO2 ranged from -22.6 ‰ to -24.3 ‰ with an average value of -23.3 ‰.
The .13CDIC in groundwater
From the range of pH values it can be deduced that bi­carbonate (HCO3-) is the dominant DIC species of the groundwater in the catchment. Therefore, concentrations of total inorganic carbon are expressed as HCO3- in this article. As shown in Tab. 2, the .13CDIC of groundwater varied from -12.4 ‰ to -10.1 ‰ in different months, with a mean value of -11.2 ‰ in 2013 in the catchment. Also, as shown in Fig. 2, obvious positive .13CDIC values of groundwater were observed in the rainy season and fertilizing period, and negative .13CDIC values of ground­water were observed in the winter, suggesting the .13CDIC of groundwater was controlled by different geochemical processes in different seasons in the QURS.
Chemical characteristics of groundwater and rainwater
As shown in Tab. 2, the pH values of groundwater samples in the QURS ranged from 7.4 to 8.1, av­eraging 7.7. The Ca2+ concentrations in ground­water varied from 103 mg/l to 131 mg/l, with a mean value of 115 mg/l. The Mg2+ concentrations varied from 9.6 mg/l to 14.5 mg/l, with a mean value of 11.9 mg/l. Ca2+ and Mg2+ dominate the cation concentrations in groundwater, accounted for 82–86 % of the total cations in groundwater. The NH4+ was not detected in the groundwater samples. HCO3- was the most abundant anions, and its concentrations ranged from 314.0 mg/l to 382.5 mg/l, averaging 345.2 mg/l. The SO42- and NO3- concentrations varied dramatically, rang­ing from 23.1 mg/l to 59.1 mg/l and 12.3 mg/l to 38.2 mg/l in different months, with a mean con­centration of 42.0 mg/l and 23.7 mg/l, respective­ly. The Cl- ranged from 14.3 mg/l to 18.5 mg/l, with a mean concentration of 16.6 mg/l.
Rainwater showed lower pH values, ranging from 4.9 to 5.1, typically characterized by acid rain. Concentrations of Ca2+, SO42- and NO3- in rainwater ranged from 11 mg/l to 12 mg/l, 11.6 mg/l to 12.6 mg/l and 6.6 mg/l to 7.5 mg/l, re­spectively. And other ions concentrations were very low in rainwater.
Meanwhile, as shown in Fig. 2, pH showed lower values in rainy season and higher values in the winter, indicating pH was impacted by the dilution effect of rainwater with lower pH in the rainy season. In contrast, the concentrations of Ca2+, Mg2+, HCO3-, NO3- and SO42- in groundwater tended to be enriched in rainy season, and the dilution effect of rainfall has not been observed in these ions of groundwater, suggesting these ions were impacted by some geochemical pro­cesses which are different from other seasons. 
Relationship between (Ca2++Mg2+) and HCO3- in groundwater
As shown in Figure 3a, the molar ratio be­tween (Ca2++Mg2+) and HCO3- of groundwater in the QURS varied from 0.56 to 0.63 with a mean value of 0.60 in different months, which higher molar ratio between (Ca2++Mg2+) and HCO3- of groundwater were observed in the rainy season and fertilizing period. As indicated by Equation (1), the molar ratio between (Ca2++Mg2+) and HCO3- released into the groundwater should be 0.5, suggesting that carbonate dissolution is con­trolled by natural processes involving carbonic acid. However, as indicated by Equation (4), (5) and (6), the molar ratio between (Ca2++Mg2+) and HCO3- released into the groundwater should be 1, suggesting that car­bonate dissolution is governed by sulphuric/nitric acid. These suggested that carbonic acid could probably not be a unique weathering agent in the catchment. Mean­while, as stated in section 4.3, higher concentrations of Ca2+, Mg2+, HCO3-, NO3- and SO42- of groundwater were observed the rainy season and fertilizing period, sug­gesting that among the different sources of protons mentioned in the introduction, fertilizers as well as at­mospheric acid deposition may play an important role in the carbonate weathering in the catchment. Therefore, the questions arised whether those elements in ground­water originated from carbonate dissolution by carbonic and sulphuric/nitric acids and to which extent of car­bonate dissolution were influenced by different acids in different months. 

Tab. 1: Isotopic composition (.13C) of limestone, plants and soil CO2 in QURS.
Sample type
 .13C (‰)
 
Limestone
 Jialingjiang formation of Lower Trissic limestone (T1j)
Leikoupo Formation of Middle Triassic limestone (T2l)
 0.2
0.3
 
Plant
 Rice
 -28.6
 
Bamboo
 -28.4
 
Cyclobalanopsis
 -26.6
 
Soil CO2
 Soil CO2 in a Rice field
 -24.3
 
Soil CO2 in a Bamboo field
 -23.1
 
Soil CO2 in a Cyclobalanopsis field
 -22.6
 



Yu Chen & Yongjun Jiang

Tab. 2: Major ions concentration and .13CDIC of groundwater and rainwater in the QURS.
Month
 Rainfall
mm
 pH
 EC
µS/cm
 Ca2+
mg/l
 Mg2+
mg/l
 Na+
mg/l
 K+
mg/l
 HCO3-
mg/l
 Cl-
mg/l
 SO42-
mg/l
 NO3-
mg/l
 NH4+
mg/l
 DO
mg/l
 .13CDIC
‰
 Contribution of carbonate dissolution by carbonic acid to total DIC in groundwater %
 Contribution of carbonate dissolution by nitric acid to total DIC in groundwater %
 Contribution of carbonate dissolution by sulphuric acid to total DIC in groundwater %
 Theoretical .13CDIC
‰
 
Groundwater
 
1
 16.0
 7.9
 596
 107
 9.60
 9.4
 5.6
 336.0
 14.5
 23.1
 12.5
 Nd
 6.2
 -12.3
 88.8
 3.7
 7.5
 -12.4
 
2
 14.6
 8.0
 605
 104
 9.8
 9.6
 5.5
 324.9
 14.6
 25.1
 14.4
 Nd
 6.4
 -12.1
 87.4
 4.4
 8.2
 -12.2
 
3
 23.8
 8.1
 624
 105
 9.6
 9.9
 5.7
 332.3
 14.4
 25.2
 15.5
 Nd
 6.6
 -12.4
 89.1
 4.6
 6.3
 -12.5
 
4
 99.0
 7.8
 616
 110
 12.1
 11.6
 5.3
 314.0
 16.4
 43.1
 25.2
 Nd
 6.8
 -10.1
 73.3
 7.9
 18.8
 -10.3
 
5
 108.6
 7.6
 633
 117
 11.3
 11.3
 5.1
 327.2
 14.3
 56.3
 35.5
 Nd
 6.8
 -10.2
 73.2
 10.7
 16.1
 -10.2
 
6
 106.3
 7.4
 678
 131
 13.5
 18.2
 3.4
 382.5
 18.5
 59.1
 38.2
 Nd
 7.1
 -10.7
 77.4
 9.8
 12.7
 -10.8
 
7
 95.2
 7.5
 664
 131
 14.5
 15.3
 4.2
 378.6
 18.1
 58.9
 37.7
 Nd
 7.3
 -10.4
 75.4
 9.8
 14.8
 -10.6
 
8
 60.4
 7.4
 655
 126
 13.6
 14.5
 3.6
 367.4
 18.2
 56.3
 35.7
 Nd
 7.2
 -10.6
 76.6
 9.6
 13.8
 -10.7
 
9
 136.0
 7.7
 647
 118
 12.3
 13.7
 3.3
 355.5
 17.9
 44.5
 22.5
 Nd
 7.1
 -11.2
 81.5
 6.2
 12.3
 -11.4
 
10
 76.3
 7.6
 643
 114
 12.4
 13.5
 3.3
 350.6
 17.6
 41.7
 19.8
 Nd
 6.9
 -11.3
 83.1
 5.6
 11.3
 -11.6
 
11
 38.9
 7.7
 633
 115
 12.1
 13.2
 3.2
 349.6
 17.5
 38.2
 14.9
 Nd
 6.8
 -11.4
 82.5
 4.2
 13.3
 -11.6
 
12
 7.0
 7.9
 593
 103
 11.8
 12.3
 3.4
 323.8
 17.3
 32.6
 12.3
 Nd
 6.2
 -11.6
 84.1
 3.7
 12.2
 -11.8
 
Rainwater
 
5
 4.9
 190
 11
 0.4
 0.3
 0.9
 12.4
 1.70
 12.6
 6.8
 0.58
 
6
 5.1
 153
 12
 0.3
 0.2
 0.7
 11.5
 1.80
 12.4
 6.6
 0.54
 
7
 5.0
 145
 12
 0.3
 0.2
 0.7
 15.4
 1.70
 11.6
 7.5
 0.49
 



The effects of agricultural activities and atmospheric acid deposition on carbonate ...




Fig. 3: Cross plot of (Ca2++Mg2+) vs. HCO3- (a) and (Ca2++Mg2+) vs. (HCO3-+NO3-+SO42-) (b) in groundwater.

Fig. 2: Seasonal variations of rainfall and hydrogeochemistry of groundwater in 2013.

Discussion

Atmospheric inputs
Compared to groundwater, the rainwater showed very low concentrations of Ca2+, Mg2+ and HCO3-, however, high concentrations of NO3- and SO42- in the catchment. Therefore, it might be concluded that Ca2+, Mg2+ and HCO3- supplied by atmospheric inputs are low, but nitrate and sulfate inputs are significant. The highest values of nitrate and sulfate were observed during local fertilizer spreading period, suggesting that they partly originate from fertilizer inputs. Meanwhile, a significant amount of NH4+ is supplied to the catchment by rainwater, but NH4+ was not detected in groundwater, which could be presumably nitrified in the catchment by soil bacteria and used by plants.
Hydrochemistry of groundwater: carbonate dissolution by carbonic 
acid versus sulphuric and nitric sulphuric acids
As atmospheric inputs of Ca2+, Mg2+ and HCO3- are very low, and the silicate weathering originated from sand­stone with slow dissolution can be neglected, the concen­trations of Ca2+, Mg2+ and HCO3- were mainly originated from carbonate dissolution in the catchment. If the con­tributions of carbonic, sulphuric and nitric acids in equi­molar amounts to the carbonate dissolution were consid­ered, carbonate dissolution becomes (Eq. 7):
4(Ca1-xMgx)CO3 + H2CO3 + HNO3 + H2SO4›
4(1-x)Ca2+ + 4xMg2+ + 5HCO3- + NO3- + SO42-	7)
In this case, the molar ratio between (Ca2++Mg2+) and HCO3-, and (Ca2++Mg2+) and (HCO3-+NO3-+SO02-) in groundwater should be 4/5 (0.8) and 4/7 (0.571). How­ever, the molar ratio between (Ca2++Mg2+) and HCO3-, (Ca2++Mg2+) and (HCO3-+NO3-+SO42-) of groundwater in different varied from 0.56 to 0.63 with a mean value of 0.60, and from 0.51 to 0.54 with a mean value of 0.52 (Figs. 3a and 3b), which deviate from the expected the molar ratio of 0.8 and 0.57, respectively, indicating car­bonate dissolution in those groundwaters controlled by carbonic, sulphuric and nitric acids in different ratios. Meanwhile, it could be assumed that all NO3- in ground­water could be derived from the nitrification of fertilizer and atmospheric NH4+, and oxidation of atmospheric NOx, due to other NO3- sources in the catchment. 
Thus, based on the (1), (5) and (6), the concentra­tions of [Ca2++Mg2+] and HCO3- in groundwater resulted from the carbonate dissolution by carbonic, nitric and sulphuric acids can be calculated by following equations:
[Ca2++Mg2+] carbonate dissolution by carbonic acid = 
1/2[HCO3-] carbonate dissolution by carbonic acid = 	(8)
[(HCO3-) groundwater - (Ca2++Mg2+)] groundwater	
[Ca2++Mg2+] carbonate dissolution by nitric acid = 
[HCO3-] carbonate dissolution by nitric acid = [NO3-] groundwater	(9)
[Ca2++Mg2+] carbonate dissolution by sulphuric acid = 
[HCO3-] carbonate dissolution by sulphuric acid = 	(10)
2(Ca2++Mg2+) groundwater - (HCO3-) groundwater - 
(NO3-) groundwater	
The calculated results are presented in Tab. 2. The contributions of carbonate dissolution by carbonic acid to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months varied from 57.7 % to 80.4 % with mean percentage of 68.5 %, and from 73.2 % to 89.1 % with a mean percentage of 81 %, re­spectively. While the contributions of carbonate dissolu­tion by nitric and sulphuric acids to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months varied from 6.4 % to 16.8 % with a mean per­centage of 11.1 %, and 3.7 % to 10.7 % with a mean per­centage of 6.7 %, and from 11.3 % to 29.7 % with a mean percentage of 20.4 % and 6.3 % to 18.8 % with a mean percentage of 12.3 %, respectively. Especially, the contri­butions of carbonate dissolution by carbonic acid to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater averaged only 63 % and 77 %, while the contributions of carbonate dissolution by nitric and sulphuric acids to to­tal concentrations of (Ca2++Mg2+) and HCO3- of ground­water averaged 37 % and 23 % in rainy season and fertil­izing period (from Apr. to Oct.). Therefore, based on the hydrochemical data, it can be concluded that carbonate dissolution was not only controlled by carbonic acid, but also by sulphuric and nitric acids introduced by the use of N-fertilizers and atmospheric acid deposition in the catchment.
The .13CDIC in groundwater: carbonate dissolution by carbonic acid versus nitric and sulphuric acids
Because the contribution of atmospheric CO2 to the DIC of groundwater was minor due to the high partial pres­sure of CO2 in groundwater, the DIC of groundwater in the catchment had two primary sources, soil CO2 and carbonate bedrock. The .13C of soil CO2 and limestone averaged -23.3 ‰ and 0.25 ‰, respectively. Meanwhile, groundwater showed higher DO (Dissolved oxygen) concentrations in different months, indicating it is an open karst system. In general, the fractionation of the carbon isotope composition was around +9 ‰ between soil gas CO2 and HCO3- (Deines 2004; Zhang et al. 1995). Therefore, DIC in groundwater originated from carbon­ate dissolving by CO2 in the catchment was expected to have .13CDIC value of around -14 ‰. While the .13CDIC had identical values to the .13C of carbonate (0.25 ‰) which DIC was derived from carbonate dissolving by sulfur acid or nitric acid. Thus, the .13CDIC in ground­water approaching a value of -14 ‰ indicated the DIC resulted from carbonate dissolving by soil CO2, while the .13CDIC in groundwater varying from 0 ‰ to -14 ‰ sug­gested the DIC resulted from carbonate dissolving by soil CO2, HNO3 and H2SO4 in the catchment. As shown in Tab. 2, the .13CDIC of groundwater varied from -12.4 ‰ to -10.1 ‰ in different months, which deviate from the expected the .13CDIC (-14 ‰) of groundwater which DIC is derived from carbonate dissolving by natural soil CO2, suggesting that the DIC of groundwater not only resulted from carbonate dissolving by natural soil CO2, but also from carbonate dissolving by sulfur and nitric acids in the catchment in the QURS.
Thus, given the .13Csoil CO2 (-23.3 ‰), .13Ccarbonate (0.25 ‰) and fractionation of isotopic compositions (+9 ‰) in groundwater system, the theoretical .13CDIC value of groundwater that DIC was derived from differ­ent sources (carbonate dissolution and soil CO2) can be calculated by the following equation:

	(11)
where .13CDIC is the theoretical .13CDIC value of ground­water, mCi is the molality of added DIC from the ith source (carbonate dissolution by carbonic acid and sul­phuric/nitric acids, respectively), and the .13Ci is .13C composition of added DIC from the ith source (soil CO2 and carbonate). 
The calculated results of the theoretical .13CDIC val­ues of groundwater are shown in Tab. 2. The measured .13CDIC value of monthly groundwater sample is much closed to its corresponding calculated .13CDIC value. Thus, the elevated .13CDIC (varying from -11.1 ‰ to -10.3 ‰ with an average of -10.6 ‰) of groundwater in the rainy season and fertilizing period suggested that carbonate dissolving was significantly influenced by nitric and sul­phuric acids originated from the use of N-fertilizers and atmospheric acid deposition in the catchment.
As indicated in Fig. 2, the .13CDIC increased with DIC in groundwater collected from the rainy season and fertilizing period, indicating that the use of N-fertilizers and atmospheric acid deposition could be responsible for the elevated DIC concentrations and .13CDIC of ground­water in the catchment. Meanwhile, an intense posi­tive relationship between the .13CDIC and (Ca2++Mg2+)/HCO3- was observed (Fig. 4). The .13CDIC increased obvi­ously with (Ca2++Mg2+)/HCO3- in groundwater collected from the rainy season and fertilizing period, indicating that the use of N-fertilizers and atmospheric acid depo­sition could be responsible for the elevated the .13CDIC and the molar ratio of (Ca2++Mg2+)/HCO3- of groundwa­ter. Also, as indicated by Fig. 4, the .13CDIC varying from -10 ‰ to -12.5 ‰, with a variational molar ratio be­tween (Ca2++Mg2+) and HCO3- of 0.53 to 0.63 of ground­water in different months, indicated the carbonate was dissolved by soil CO2 (from C3 vegetation), HNO3 and H2SO4 in the catchment. Thus, the evidences from both of the chemical data and .13CDIC of groundwater in dif­ferent months indicated that carbonate dissolution was controlled by natural carbonic acid, and nitric and sul­phuric acids originated from the use of N-fertilizers and atmospheric acid deposition in the QURS.

Yu Chen & Yongjun Jiang

The effects of agricultural activities and atmospheric acid deposition on carbonate ...


Fig. 4: Cross plot of (a) .13CDIC vs. (Ca2++Mg2+)/HCO3- of ground­water in different seasons.

Conclusions

Although the results of this study are preliminary and the effects of the use of N-fertilizers and acid depo­sition on carbonate weathering have been quantified in a small karstic agricultural catchment of Southwest China.
The significant temporal variations for major ele­ment concentrations and .13CDIC of groundwater in dif­ferent months were observed in the QURS. The ground­water collected in rainy season and fertilizing period showed significant high concentrations of DIC, Ca2+, Mg2+, NO3-, SO42- and .13CDIC. These temporal variations could be attributed to carbonate dissolution by natural carbonic acid, and nitric and sulphuric acids introduced by the use of N-fertilizers and atmospheric acid deposi­tion in the QURS. 
The contributions of carbonate dissolution by car­bonic acid to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months averaged 68.5 % and 81 %, respectively. While the contributions of carbonate dissolution by nitric and sulphuric acids originated from the use of N-fertilizers and atmospheric acid deposition to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater in different months averaged 11.1 % and 6.7 %, and 20.4 % and 12.3 %, respectively. Especially, the contributions of carbonate dissolution by carbonic acid to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater averaged only 63 % and 77 %, while the contributions of carbonate dissolution by nitric and sulphuric acids to total concentrations of (Ca2++Mg2+) and HCO3- of groundwater averaged 37 % and 23 % in rainy season and fertilizing period (from Apr. to Oct.). 
The temporal variations of .13CDIC in groundwater (varying from -12.4 ‰ to -10.1 ‰) in different months deviated from the expected the .13CDIC (-14 ‰) of groundwater which DIC is derived from carbonate dis­solving by natural soil CO2, suggesting that the DIC of groundwater not only resulted from carbonate dissolving by natural soil CO2, but also from carbonate dissolving by sulfur and nitric acids in the QURS. Meanwhile, the .13CDIC increased obviously with (Ca2++Mg2+)/HCO3- of groundwater in the rainy season and fertilizing period, indicating that the use of N-fertilizers and atmospheric acid deposition should be responsible for the elevated the .13CDIC and the molar ratio of (Ca2++Mg2+)/HCO3- of groundwater in the QURS. 
Thus, the evidences from both of the hydrochemical data and .13CDIC of groundwater in different months indi­cated that carbonate dissolution was not only controlled by natural carbonic acid, but also by sulphuric and nitric acids introduced by the use of N-fertilizers and atmo­spheric acid deposition in the QURS. More important is that not only the concentrations of nitrate and sulphate in karst groundwater have been elevated, but also the exports of inorganic carbon and (Ca2++Mg2+) have been enhanced due to carbonate weathering by nitric and sul­phuric acids originated from the use of N-fertilizers and atmospheric acid deposition in the catchment.

Yu Chen & Yongjun Jiang

Acknowledgments

This work is financially supported by National Natural Science Foundation of China (Grant Nos. 41472321 and 41172331), and IGCP598.

References

Ali, H.N. & E.A. Atekwana, 2011: The effect of sulfuric acid neutralization on carbonate and stable car­bon isotope evolution of shallow groundwater.- Chemical Geology, 284, 217-228. http://dx.doi.org/10.1016/j.chemgeo.2011.02.023
Atekwana, E.A. & R.V. Krishnamurthy, 1998: Seasonal variations of dissolved inorganic carbon and .13C of surface waters: application of a modified gas evolution technique.- Journal of Hydrology, 205, 265-278. http://dx.doi.org/10.1016/S0022-1694­(98)00080-8
Barnes, R.T. & P.A. Raymond, 2009: The contribution of agricultural and urban activities to inorganic car­bon fluxes within temperate watersheds.- Chemi­cal Geology, 266, 3-4, 327-336. http://dx.doi.org/10.1016/j.chemgeo.2009.06.018
Calmels, D., Gaillardet, J., Brenot, A. & C. France-Lan­ord, 2007: Sustained sulfide oxidation by physical erosion processes in the Mackenzie River basin: Climatic perspectives.- Geology, 35, 11, 1003-1006. http://dx.doi.org/10.1130/G24132A.1
Deines, P., 2004: Carbon isotope effects in carbon­ate systems.- Geochimica et Cosmochimica Acta, 68, 2659-2679. http://dx.doi.org/10.1016/j.gca.2003.12.002
Dentener, F., Drevet, J., Lamarque, J.F., Bey, I., Eick­hout, B., Fiore, A.M., Hauglustaine, D., Horowitz, L.W., Krol, M., Kulshrestha, U.C., Lawrence, M., Galy-Lacaux, C., Rast, S., Shindell, D., Stevenson, D., Noije, T. van, Atherton, C., Bell, N., Bergman, D., Butler, T., Cofala, J., Collins, B., Doherty, R., El­lingsen, K., Galloway, J., Gauss, M., Montanaro, V., Müller, J.F., Pitari, G., Rodriguez, J., Sanderson, M., Solmon, F., Strahan, S., Schultz, M., Sudo, K., Szopa S. & O. Wild, 2006: Nitrogen and sulfur deposition on regional and global scales: A multimodel evalu­ation.- Global Biogeochemical Cycles, 20, GB4003, 1-21. http://dx.doi.org/10.1029/2005GB002672
Gandois, L., Perrin, A.-S. & A. Probst, 2011: Impact of nitrogenous fertiliser-induced proton release on cultivated soils with contrasting carbonate contents: a column experiment.- Geochimica et Cosmochim­ica Acta, 75, 1185-1198. http://dx.doi.org/10.1016/j.gca.2010.11.025
He, Q., Yang P., Yuan, W., Jiang, Y., Pu, J., Yuan, D. & Y. Kuang, 2010: The use of nitrate, bacteria and fluo­rescent tracers to characterize groundwater recharge and contamination in a karst catchment, Chongq­ing, China.- Hydrogeology Journal, 18, 1281-1289. http://dx.doi.org/ 10.1007/s10040-010-0594-0
Jiang, Y., 2012: Sources of sulfur in the Nandong under­ground river system, southwest China: A chemical and isotopic reconnaissance.- Applied Geochemis­try, 27, 8, 1463-1470. http://dx.doi.org/10.1016/j.apgeochem.2012.05.001
Jiang, Y., 2013: The contribution of human activities to dissolved inorganic carbon fluxes in a karst under­ground river system: evidence from major elements and .13CDIC in Nandong, Southwest China.- Jour­nal of Contaminant Hydrology, 152, 1-11. http://dx.doi.org/10.1016/j.jconhyd.2013.05.010.
Jiang, Y., Wu, Y., Groves, C., Yuan, D. & P. Kambesis, 2009a: Natural and anthropogenic factors affect­ing the groundwater quality in the Nandong karst underground river system in Yunan, China.- Jour­nal of Contaminant Hydrology, 109, 49-61. http://dx.doi.org/10.1016/j.jconhyd.2009.08.001
Jiang, Y., Wu, Y. & D. Yuan, 2009b: Human impacts on karst groundwater contamination deduced by coupled nitrogen with strontium isotopes in the Nandong underground river system in Yunan, China.- Environmental Science & Technology, 43, 7676-7683. http://dx.doi.org/10.1021/es901502t
Jiang, Y., Yuan, D., Zhang, C., Zhang, G. & R. He, 2008: Impact of land use change on groundwater qual­ity in a typical karst watershed of southwest Chi­na.- Hydrogeology Journal, 16, 4, 727–735. http://dx.doi.org/10.1007/s10040-007-0259-9
Li, S., Calmel, D., Han, G., Gaillardet, J. & C. Liu, 2008: Sulfuric acid as an agent of carbonate weather­ing constrained by .13CDIC: Examples from South­west China.- Earth and Planetary Science Let­ters, 270, 189-199. http://dx.doi.org/10.1016/j.epsl.2008.02.039
Li, S., Liu, C., Li, J., Lang, Y., Ding, H. & L. Li, 2010: Geochemistry of dissolved inorganic carbon and carbonate weathering in a small typical karstic catchment of Southwest China: Isotopic and chemi­cal constraints.- Chemical Geology, 277, 301-309. http://dx.doi.org/10.1016/j.chemgeo.2010.08.013
Liu, C., Li, S., Lang, Y. & H. Xiao, 2006: Using .15N- and .18O-values to identify nitrate sources in karst ground water, Guiyang, southwest China.- Environ­mental Science & Technology, 40, 22, 6928-6933. http://dx.doi.org/10.1021/es0610129
Liu, C., Lang, Y., Satake, H., Wu, J. & S. Li, 2008: Iden­tification of anthropogenic and natural inputs of sulfate and chloride into the karstic ground water of Guiyang, SW China: Combined .37Cl and .34S approach.- Environmental Science & Technol­ogy, 42, 15, 5421-5427. http://dx.doi.org/10.1021/es800380w
Paulot, F., Jacob, D.J. & D.K. Henze, 2013: Sources and processes contributing to nitrogen deposition: an adjoint model analysis applied to biodiver­sity hotspots worldwide.- Environmental Science & Technology, 47, 7, 3226-3233. http://dx.doi.org/10.1021/es3027727
Perrin, A.-S., Probst, A. & J.-L. Probst, 2008: Impact of nitrogenous fertilizers on carbonate dissolu­tion in small agricultural catchments: implica­tions for weathering CO2 uptake at regional and global scales.- Geochimica et Cosmochimica Acta, 72, 3105-3123. http://dx.doi.org/10.1016/j.gca.2008.04.011
Pu, J., Yuan, D., He, Q., Wang, Z., Hu, Z. & P. Gou, 2011: High-resolution monitoring of nitrate variations in a typical subterranean karst stream, Chongq­ing, China.- Environmental Earth Science, 64, 7, 1985-1993. http://dx.doi.org/10.1007/s12665-011-1019-7
Semhi, K., Suchet, P. A., Clauer, N. & J.-L. Probst, 2000: Impact of nitrogen fertilizers on the natural weath­ering-erosion processes and fluvial transport in the Garonne basin.- Applied Geochemistry, 15, 6, 865-874. http://dx.doi.org/10.1016/S0883-2927­(99)00076-1
Tang, J., Xu, X., Ba J. & S. Wang, 2010: Trends of the pre­cipitation acidity over China during 1992-2006.- Chinese Science Bulletin, 55, 17, 1800-1807. http://dx.doi.org/s11434-009-3618-1
Yue F.-J., Li S.-L., Liu C.-Q., Lang Y.-C. & H. Ding, 2015: Sources and transport of nitrate constrained by the isotopic technique in a karst catchment: an example from Southwest China.- Hydrological Processes, 29, 8, 1883-1893. http://dx.doi.org10.1002/hyp.10302/
Zhang, J., Quay, P.D. & D.O. Wilbur, 1995: Carbon iso­tope fractionation during gas-water exchange and dissolution of CO2.- Geochimica et Cosmochimica Acta, 59, 107-114. http://dx.doi.org/10.1016/0016-7037(95)91550-D

The effects of agricultural activities and atmospheric acid deposition on carbonate ...

Yu Chen & Yongjun Jiang