}northeastern.edu/protect Characterization of Fate and Transport Processes and Contaminant Distribution in Karst Groundwater Systems Norma I. Torres ([email protected]), Jonathan Toro-Vázquez ([email protected]), Elienisse Rodriguez ([email protected]), and Ingrid Y. Padilla ([email protected]) Department of Civil Engineering and Surveying, University of Puerto Rico at Mayagüez Introduction Objectives References Methodology Preliminary Results Study area Puerto Rico Northern Karst Aquifer system (Fig. 3a): Fig. 3. (a) Hydrogeology of Puerto Rico (b) Historical groundwater contamination in the North Coast of PR 1 (a) (b) A B C Previous studies in the northern karst aquifers of Puerto Rico show significant distribution of contaminants beyond demarked sources of contamination 1 (Fig. 3b). Most extensive and productive aquifer of the island 6 . Affected by a long history of toxic spills, chemical waste and industrial solvent release into the subsurface 7 . Contaminated sites has been under active remediation during the past decades 7 . Chlorinated Volatile Organic Compounds (CVOCs) 7 Solvents Degreasers Paint Removers Study focuses on CVOCs (Fig. 4) Characterize of fate and transport processes in karst groundwater systems at laboratory and field-scales. Assess spatial and temporal contaminant distribution of contaminants in karst groundwater systems (regional scale). Presence in the environment Presence in listed superfund sites in PR and the US Potential exposure and human health problems Water quality records of regulatory agencies (2011-2015) Current sampling and analysis (1981-2015) GIS: spatial and temporal maps of attributes Detection maps Concentrations distribution maps of total CVOCs Total CVOCs=sum of the concentrations of the detected CVOCs in the study area Proximity analysis Descriptive analysis, detection frequencies per site, sample, and contaminant, temporal distribution of contaminants ANOVA Chi-Square Test Logistic regression models Tracer test study: Rhodamine and Uranine 2.0 g of Rhodamine and 3.2 g of Uranine were injected in “El Tallonal Cave”(Fig. 8). Fig. 8. Location of the sampling points in the Tallonal Cave Incomplete recovery of tarcers (Fig. 9). Method of moments was used to determine the coefficient of dispersion (D) between the three points. Karst aquifers: Highly productive aquifers 1 : characterized by springs, caves, sinkholes, interconnected fissures, fractures and conduits 2 (Fig. 1). Vulnerable to contamination: high capacity to store and convey contaminants to zones of potential exposure 1 (Fig. 2). Contamination may be influenced by anthropogenic and/or hydrogeological factors 5 . Fig. 1. Cross section of a karst aquifer 3 Fig. 2. Groundwater flow in a karst system 4 High heterogeneity and anisotropy: Prevents accurate prediction in contaminant fate and transport. Challenges in understanding the impacts of hydrologic conditions changes on fate and transport processes. Limited technologies to characterize and quantify flow and transport processes at field-scale. Dispersion among tracers varies with distance and flow rate (Fig. 7): Dispersion values tend to increase with distance for both uranine and rhodamine wt. Uranine and rhodamine dispersion tend to increase for high flow rates. Rhodamine wt dispersion values are slightly higher than for uranine. cm 2 /min Rhodamine WT Base Flow = 0.25 GPM Base Flow = 0.5 GPM Base Flow = 1 GPM Uranine Base Flow = 0.25 GPM Base Flow = 0.5 GPM Base Flow = 1 GPM Limestone Length (cm) Limestone Height (cm) 145 cm 59 cm 59 cm Karstified Limestone Physical Model (KLPM) (Fig. 5): Rhodamine and Uranine tracers tests under several flow conditions (Fig. 6.). Fig. 5. Illustration of the KLPM 3 4 1 2 7 8 5 6 9 12 11 10 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 IN OUT Fig. 6. Scheme of the ports on the KLPM Spatio-temporal concentration distribution (STCD) developed using Golden Software Surfer v12. Fate and transport parameters were estimated for multi-tracers using CXTFIT code 8 (Equation 1). Equation 1. Dimensionless Mobile-Immobile Equation 0 1000 2000 3000 4000 5000 6000 0 500 1000 1500 Concentration (ppb) Time (sec.) Concentration of tracers in point A 0 200 400 600 800 1000 0 500 1000 1500 Concentration (ppb) Time (sec.) Concentration of tracers in point B 0 50 100 150 200 250 300 350 400 -400 100 600 1100 1600 Concentration (ppb) Time (sec.) Concentration of tracers in point C Rhodamine Uranine Fig. 11 Detection frequencies and concentrations of most detected CVOCs. CVOCs were detected as single entities or mixtures in 64% of the samples and 77% of the sites sampled (Fig. 10). Most frequently detected CVOCs include: TCE, PCE, TCM, cis-1,2-DCE, 1,1-DCE and CCl 4 , with average concentrations ranging from 0.0045 to 0.1203 mg/L (Fig. 11). Results from the Moment analysis for the coefficient of dispersion: Laboratory Scale Methodology Preliminary Results Points B-A C-B C-A D (cm 2 /min) 1.62 4.14 3.22 Rhodamine Uranine Points B-A C-B C-A D (cm 2 /min) 0.94 3.83 2.88 Analysis showed that CVOCs total concentrations are significantly higher in: Wells located in the upper aquifer, within areas of low sinkhole coverage and low hydraulic conductivities, and wet season Wells within a distance of 0 to 3.2 km from superfund sites, 3.2 to 6.4 km from RCRA- CA sites, and more than 6.4 km from landfills. Fig. 10. Detection of CVOCs in the study area Analysis showed that detection of CVOCs are significantly higher in: Wells located in the upper aquifer, within areas of low and intermediate sinkhole coverage, intermediate and high hydraulic conductivities, and dry season (hydrogeological factors) Wells within a distance of 0 to 3.2 km from superfund sites, 3.2 to 6.4 km from RCRA-CA, and 3.2 to 6.4 km from landfills (anthropogenic factors). Logistic regression model indicates that: The detection of CVOCs in the karsts aquifers of NPR is influenced by a combination of contaminant source and hydrogeological factors. Conclusions Total (average) CVOCs concentration (mg/L) Spatial distribution of concentrations showed an extensive spatial groundwater contamination with CVOCs from multiple sources (Fig. 12). Fig. 12. Total CVOCs concentrations distribution in the study area At the lab scale, the spatial distribution of the estimated fate and transport parameters for the tracers revealed high variability related to preferential flow heterogeneities and scale dependence. Field scale and lab-scale tracer analysis showed differences in fate and transport parameters that result from different system conditions at both scales (flow, dimensions, etc.) The regional scale analysis of contaminant distribution suggested that CVOCs are persistent contaminants in karst systems, even under active remediation, they are present in groundwater for more than 30 years. Detection and spatial distribution of CVOCs are influenced by the type of source of contamination, and the characteristics of the karst system. Adequate characterization of flow, fate, and transport processes in karst aquifers will enhance the ability to predict and minimized the potential for exposure of contaminants to humans and ecosystems, and provide tools for better decisions on the management of groundwater systems. (1) Padilla, I.Y., C. Irizarry, and K. Steele. (2011). Historical Contamination of Groundwater Resources in The North Coast Karst Aquifers Of Puerto Rico, Dimensión, Año 25 Vol 3, 7-12, 2011. (2) United States Geological Survey (USGS). (2012). USGS Groundwater Information. Retrieved January 13, 2013, from http://water.usgs.gov/ogw/karst/pages/whatiskarst. (3) Physical Geology . (nd). Laboratory 12-Groundwater Processes, Resources, and Risks. Retrieved November 10, 2015 from http://www.ocean.odu.edu/~spars001/physical_geology/laboratory/laboratory_12/handout_laboratory_12.html (4) United Nations Environment Programme. (nd). Groundwater: Water Flowing Under the Land Surface. Retrieved November 10, 2015 from http://www.unep.or.jp/ietc/Publications/Short_Series/LakeReservoirs-2/9.asp (5) Torres, N.I., Yu, X., Padilla, I.Y., Macchiavelli, R.E., Ghasemizadeh, R., Kaeli, D., Cordero, J.F., Meeker, J.D., Alshawabkeh, A.N. (2018). The influence of hydrogeological and anthropogenic variables on phthalae contamination in eogenetic karst groundwater systems. Environmental Pollution, 237, 298-307. (6) Lugo A.E., Miranda Castro, L., Vale, A., López, T. Del M., Hernández Prieto, E., García Martinó, A., Puente Rolón, A.R., Tossas, A.G., McFarlane, D.A., Miller, T., Rodríguez, A., Lundberg, J., Thomlinson, J., Colón, J., Schellekens, J.H., Ramos, O., and Helmer, E. (2001). Puerto Rican Karst-A Vital Resource, U.S. Forest Service Gen. Tech. Report WO-65. (7) Padilla I.Y., Rivera, V.L., and Irizarry, C. (2015). Spatiotemporal Response of CVOC Contamination and Remedial Actions in Eogenetic Karst Aquifers. Proceedings of the 14 th Multidisciplinary Conference on Sinkholes and the Engineering and Environmental Impacts of Karst, October 5-9, 2015. Rochester, Minnesota, pp. 337-345. (8) Moran, M. J., Zogoroski, J. S., and Squillace, P. J. (2007). Chlorinated Solvents in Groundwater of the United States. Environmental Science and Technology , 41, 74-81. (9) Tang, G., M.A. Mayes, J.C. Parker and P.M. Jardine. (2010). CXTFIT/Excel-a modular adaptable code for parameter estimation, sensitivity analysis, and uncertainty analysis for laboratory and field tracer experiments . Computers & Geosciences. 36(9): 1200- 1209. DOI:10.1016/j.cageo.2010.01.013 Fig. 4. (a) Importance of studying CVOCs, (b) Products where are CVOCs are commonly found (a) (b) Fig. 7. Spatial distribution of the coefficient of dispersion on the KLPM. Fig. 9. Concentration of tracers in points A, B, C Preliminary Results Methodology = . = . = . = . = . = . = . = . = . Preliminary Results (cont.) Field Scale Regional Scale