TY - JOUR
T1 - Role of an impermeable layer in controlling groundwater chemistry in a basaltic aquifer beneath an agricultural field, Jeju Island, South Korea
AU - Jung, Hee Won
AU - Yun, Seong Taek
AU - Kim, Kyoung Ho
AU - Oh, Sang Sil
AU - Kang, Kyung Goo
N1 - Funding Information:
This work was supported by funds from the project, “Leading Industry Developments in the Economic Region”, provided by the Ministry of Knowledge Economy (MKE) through the Korea Institute for the Advancement of Technology (KIAT). Support was also provided by the 2012 Jeju Green Environment Center Project. Prof. Rodney Grapes helped to improve early drafts of this manuscript. Constructive comments provided by anonymous reviewers helped to clarify and improve the manuscript.
PY - 2014/6
Y1 - 2014/6
N2 - Hydrochemical data (major cations and anions, and trace metals such as Fe, Mn and V) of groundwater (n=144) were collected from 15 pre-existing wells in the Gosan area, along the southwestern coast of Jeju Island, South Korea, between January and October 2010. These were used to investigate the spatio-temporal variability of groundwater quality in a basaltic aquifer beneath an agricultural field. In the western part of the study area, an impermeable clay-rich layer (Gosan Formation) overlies the basaltic aquifer, prohibiting direct recharge from rainwater infiltration and allowing a shallow perched aquifer to form above the formation. Robust Principal Component Analysis (ROBPCA) was performed to investigate the hydrochemical characteristics. Twenty-seven outlying samples were separated from the total dataset; among these, 17 samples reflected seawater effects, and seven samples reflected abnormally high impacts from agricultural contamination. For the rest regular observations (n=117) and a good leverage point, ROBPCA showed that principal component (PC) 1 effectively distinguishes uncontaminated water from those samples reflecting agricultural contamination. Variables with positive PC 1 loadings (TDS, Ca, Mg, NO3, SO4, and Cl) indicate agricultural contamination, while negative PC 1 loadings (pH, Na, K, HCO3 and V) possibly indicate basalt weathering. Samples with chemistry controlled by agricultural contamination are restricted to the eastern part of the study area, while uncontaminated water is predominantly observed in the western part where the Gosan Formation occurs. Such a distinct spatial pattern indicates that overall groundwater quality is regulated by the occurrence of the impermeable formation. Groundwater samples from wells near the edge of the Gosan Formation show seasonal fluctuations in water quality, with factor scores that indicates uncontaminated water in the dry season (January to April), and agricultural contamination in the wet season (May to October). This suggests that groundwater below the marginal part of the impermeable layer is seasonally contaminated by a temporal extension of the pollution front during the wet season, even though the impermeable layer plays a role as a natural barrier to protect groundwater from the infiltration of surface contaminants. This study also shows the advantage of ROBPCA to successfully identify spatio-temporal variation of groundwater quality in an area where diverse hydrochemical processes are coexisting and anormalous samples occur.
AB - Hydrochemical data (major cations and anions, and trace metals such as Fe, Mn and V) of groundwater (n=144) were collected from 15 pre-existing wells in the Gosan area, along the southwestern coast of Jeju Island, South Korea, between January and October 2010. These were used to investigate the spatio-temporal variability of groundwater quality in a basaltic aquifer beneath an agricultural field. In the western part of the study area, an impermeable clay-rich layer (Gosan Formation) overlies the basaltic aquifer, prohibiting direct recharge from rainwater infiltration and allowing a shallow perched aquifer to form above the formation. Robust Principal Component Analysis (ROBPCA) was performed to investigate the hydrochemical characteristics. Twenty-seven outlying samples were separated from the total dataset; among these, 17 samples reflected seawater effects, and seven samples reflected abnormally high impacts from agricultural contamination. For the rest regular observations (n=117) and a good leverage point, ROBPCA showed that principal component (PC) 1 effectively distinguishes uncontaminated water from those samples reflecting agricultural contamination. Variables with positive PC 1 loadings (TDS, Ca, Mg, NO3, SO4, and Cl) indicate agricultural contamination, while negative PC 1 loadings (pH, Na, K, HCO3 and V) possibly indicate basalt weathering. Samples with chemistry controlled by agricultural contamination are restricted to the eastern part of the study area, while uncontaminated water is predominantly observed in the western part where the Gosan Formation occurs. Such a distinct spatial pattern indicates that overall groundwater quality is regulated by the occurrence of the impermeable formation. Groundwater samples from wells near the edge of the Gosan Formation show seasonal fluctuations in water quality, with factor scores that indicates uncontaminated water in the dry season (January to April), and agricultural contamination in the wet season (May to October). This suggests that groundwater below the marginal part of the impermeable layer is seasonally contaminated by a temporal extension of the pollution front during the wet season, even though the impermeable layer plays a role as a natural barrier to protect groundwater from the infiltration of surface contaminants. This study also shows the advantage of ROBPCA to successfully identify spatio-temporal variation of groundwater quality in an area where diverse hydrochemical processes are coexisting and anormalous samples occur.
UR - http://www.scopus.com/inward/record.url?scp=84897980717&partnerID=8YFLogxK
U2 - 10.1016/j.apgeochem.2014.03.008
DO - 10.1016/j.apgeochem.2014.03.008
M3 - Article
AN - SCOPUS:84897980717
SN - 0883-2927
VL - 45
SP - 82
EP - 93
JO - Applied Geochemistry
JF - Applied Geochemistry
ER -