Groundwater-surface water interactions in a volcanic maar lake (Hule maar, Costa Rica)

Published: 25 April 2024| Version 1 | DOI: 10.17632/s87wf8zdgf.1
Germain Esquivel-Hernandez,


These data include stable isotopes of carbon and water, radon-222 in water, hydrochemistry, and weather data. This study was carried out at Lake Hule, a volcanic lake in northern Costa Rica, and aimed to better understand the interactions between groundwater and surface water. Our results showed little difference in evaporation-to-inflow ratios between dry (December-April) and wet (May-November) seasons. Groundwater, precipitation, and runoff contributed ~61.3%, 24.4%, and 14.3% of total inflow to the lake, respectively. We found that carbonate buffering also played a key role in the lake chemistry, with greater carbon dioxide degassing from groundwater sources in the wet season. This study provides insight into groundwater-surface water interactions in Central America's volcanic front.


Steps to reproduce

Water samples of Lake Hule, springs, and streams were collected between March 2022 and September 2023. A passive collector was installed (Palmex Ltd., Croatia; Gröning et al., 2012) in July 2022 above the caldera wall on the east side of Lake Hule (Latitude: 10.2951°, Longitude: -84.2033°, 875 m asl) to manually collect on every rainy day in the morning (7-8 am) precipitation samples from July 2022 to September 2023. To better characterize the groundwater system in the lake basin, we collected water samples from three springs, namely Pata Gallo (982 m asl), Rio Cuarto (651 m asl), and Crucero (617 m asl). Samples were collected manually every week in the periods of March-October 2022 and March-September 2023. Water samples of streams, namely Maria Aguilar river (736 m asl), Sardinal river (453 m asl), Hule river (418 m asl), and Rio Cuarto river (417 m asl) were collected during sampling campaigns in March 2022, March 2023, and September 2023 . At Lake Hule, samples were also collected in March 2022, March 2023, and September 2023 at five sites located along a transect in the west-east direction (N=39, Figure 1). Lake samples were collected (~ 1 m below the surface) and bottom samples (in the last 1 m of the water column and as close as possible to the lake floor). The depth at which lake bottom samples were collected was in the range of 10-20 m. Water samples were collected using a 2.2 L Niskin bottle sampler (Wildco, USA). The stable isotope composition of lake, springs, streams, and precipitation samples was analyzed by CRDS spectroscopy (L2140-i, Picarro, USA). The stable isotope composition of dissolved inorganic carbon (δ13CDIC) and total alkalinity of filtered aliquots (0.45μm PTFE) were also analyzed by CRDS spectroscopy (G2201-i, Picarro, USA). Ion chromatography (Thermo Scientific ICS-5000+, CA, USA) was used to analyze ammonium, sodium, potassium, magnesium, calcium, chloride, nitrite, nitrate, and sulfate. The radon-222 activity was measured in lake, groundwater, and stream water samples using a RAD7 detector (Durridge Co., USA). Hourly meteorological variables (relative humidity, air temperature, solar radiation, wind speed and direction, and precipitation amount) were recorded at 2 m height using a Vantage Pro2 weather station (Davis Instruments, USA), which was installed next to the precipitation collector.


Universidad Nacional de Costa Rica, International Atomic Energy Agency


Water Isotopes, Radon, Carbon Isotope, Applied Limnology


International Atomic Energy Agency

COS- 24985