Submarine Groundwater Discharge

Submarine groundwater discharge (SGD) is groundwater leaking from a coastal aquifer into the ocean. It manifests as coastal springs or diffuse seepage and occurs anywhere from nearshore to offshore regions sometimes at great distances and depths from the coastline. SGD has great significance as a fresh and brackish water source for people and a source of nutrients and freshwater to groundwater dependent ecosystems inhabiting the coastline, fishponds, and anchialine pools.

SGD plays an important role in an island water budget – whatever goes in has to come out: recharge into an aquifer feeds coastal springs and seeps, stream baseflow, and water withdrawal for human use. The more water is taken out for human use, the less is available to feed streams and coastal springs. For proper water management, we need to know where and how much groundwater recharges into an aquifer and how different aquifers are interconnected. Chemical fingerprints of groundwater discharging at the coast can inform us where the water originates, what land-use it flows under and how much time it spends in the aquifer.

SGD schematic
Conceptual hydrogeologic model of groundwater and SGD flow. From Bishop et al. (2017)

Coastal springs discharging along the coastline of the Hualālai region from Kīholo and Keauhou aquifers were sampled to determine water quality and chemical characteristics of the groundwater. Stable isotopes of water were used to determine recharge origin and possible subsurface water flow lines. Specifically, as shown in previous studies there is a predictable relationship between ð18O of water and precipitation elevation, a relationship that allows distinguishing recharge from the slopes of Hualalai (<2500 m) and Mauna Kea and Mauna Loa (>2500 m). Cluster analysis was used to identify similarities and differences between stable isotope signatures within and between aquifers. According to this analysis, coastal springs group according to similarities in recharge elevation and resulting possible flow patterns.  The distribution and extent of these groups is illustrated on Figure 1 with different groups assigned a different color for easy identification. The major conclusions of this analysis are that groundwater in multiple groups of coastal springs have to originate beyond aquifer boundaries and that coastal springs in the Keauhou aquifer cannot be explained by recharge solely in the basal lens suggesting water contribution from the high-level aquifer.

Display of groundwater recharge lines for Hualālai, Kīholo, and Keauhou
Groundwater discharging from coastal springs can be traced back into its potential recharge location based on its water stable isotopic composition. The results suggest that in many cases spring flow depends on recharge in neighboring watersheds. Mauka to makai water management here extends over 2-3 volcanoes

A new generation of SGD Sniffer, an autonomous SGD monitor, has been developed to document SGD in fishponds and anchialine pools. Understanding the magnitude and trends in groundwater discharge to these coastal ecosystems is vital for their better management and preservation. Optimal salinity and nutrient distribution in ponds are maintained by the interplay between groundwater discharge and seawater intrusion (driven by tides & seasonal water level fluctuations). Using long-term groundwater flow observations we can determine baseline and extreme (wet/dry season and high/low sea level) conditions to understand natural variability and thresholds of minimum groundwater flow required for keeping salinity and nutrient fluxes within optimal limits.  Another parameter that this research provides is the amount of nitrogen delivered by SGD to coastal systems. Nitrogen has both natural and man-made sources and amounts beyond an optimal level have detrimental effects on native groundwater dependent ecosystems.

SGD sniffer floating in pond
One of the new SGD Sniffers deployed in a coastal pond in West Hawaiʻi.
Graph of SGD machine learning results
Machine learning can provide additional insights in to the drivers of SGD. We are applying machine learning to a long-term, high-resolution SGD Sniffer dataset to better understand the drivers of SGD and how it may change in the future with climate change and sea level rise. The SGD Sniffer captured a high sea level anomaly in the Pacific Ocean in 2017 and the model is currently able to accurately predict SGD up until the high sea level anomaly occurs. We hope to improve the model and to use this high sea level anomaly to better understand SGD in the future.

Submarine groundwater discharge is driven by the hydraulic gradient between the groundwater aquifer and the ocean. The major drivers of discharge are precipitation/recharge from the terrestrial side and ocean water level such as tides and seasonal sea level fluctuation from the marine side. A novel instrument called SGD Sniffer has been constructed to perform long-term, high frequency monitoring of SGD. Time series analysis of the SGD record as well as ocean water level and precipitation data showed that tides are the “gate keeper” for SGD allowing more groundwater to discharge during low tides. Machine learning methods are used to analyze temporal patterns of SGD and its variation with terrestrial and marine driving forces to be able to predict SGD trends under future climate and sea level rise scenarios.

You can learn more about our work here:

VOS4-1 Full Episode – Mapping the Freshwater of Hawai’i:

‘Ike Wai Summer Bridge, 2017

Datasets:

Honolulu King Tide Study: Radon Time Series: Time series results by study site and sampling date (including king tide vs. spring tide sampling for coastal sites). Data collected include time, water temperature, water salinity, water depth, and radon concentrations in water.

Honolulu King Tide Study: Raw sample dataset:  Contains all grab sample data collected, including location, date, lat long, salinity, water depth, radon concentration in water, carbamazepine concentrations, caffeine concentrations, fluoroquinolones concentrations, and dissolved nutrient concentrations (including phosphate, nitrate + nitrite, ammonium, total dissolved nitrogen, and total dissolved phosphorus).

Papers and conference abstracts:
20th EGU General Assembly, EGU2018
2018 Ocean Sciences Meeting
AGU Fall Meeting 2019
Goldshmidt 2020
Collaborative research to support urban agriculture in the face of change: The case of the Sumida watercress farm on O‘ahu

 

SGD Team members:

Henrietta Dulai -Lead, Associate Professor, Department of Earth Sciences, SOEST, UHM

Catherine Hudson – Graduate student, Department of Earth Sciences, SOEST, UHM
Trista McKenzie – Graduate student, Department of Earth Sciences, SOEST, UHM

We are grateful to our community partners for their help in making this project happen.