Groundwater Flow Modeling - Long Island, New York
Numerical models provide a means to synthesize existing hydrogeologic information into an internally consistent mathematical representation of a real system or process, and thus are useful tools for testing and improving conceptual models or hypotheses of groundwater flow systems. The goal of this effort is to develop a regional model for the Long Island aquifer system to simulate changes in water levels, streamflows, the position and movement of the boundary between fresh and saline groundwater, assess groundwater age distributions, and to determine regional hydrologic response to changes in anthropogenic and natural stresses in order to assist in the determination of the groundwater sustainability of the Long Island aquifer system.
Modeling Approach:
Pertinent hydrogeologic information is needed to develop a conceptual model of how water enters, moves through, and ultimately leaves the aquifer system. The conceptual model then is used as the foundation for the construction of a 3D-numerical model capable of simulating groundwater flow conditions throughout the Long Island aquifer system. The collection of digital-hydrogeologic data in the study area are needed for proper model development, including elevations and extents of hydrostratigraphic units, groundwater salinity distribution, groundwater withdrawals, temperature and precipitation distributions, recharge and impervious-surface distribution, and additional information necessary for accurate model calibration including groundwater levels and streamflow.
The sources (or inflows) and sinks (or outflows) of water to the Long Island aquifer system can be illustrated schematically (fig. 1) to show how water enters, flows through, and exits this aquifer system). The sole source of water for predevelopment conditions is recharge from precipitation. All of this water entering the aquifer system is balanced by water leaving the aquifer system. Water leaves the aquifer system as discharge to streams, discharge to shallow-coastal waters, and deep-subsea discharge farther offshore in the Atlantic Ocean. An additional loss of water from the aquifer system for post- development conditions is water removed by groundwater withdrawals. Depending on the type of water use, and whether (or how much) a community is sewered a substantial amount of the water withdrawn from the aquifer system may be returned in the form of wastewater-return flow through domestic septic systems.
Modeling Components:
The main components associated with the model development include (1) hydrogeologic framework, (2) aquifer recharge, (3) water use, (4) freshwater/saltwater boundary, and (5) model calibration.
Hydrogeologic Framework:
The hydrogeologic framework consists of the extents and thicknesses of the major hydrogeologic units and the hydraulic properties of the sediments that comprise these units. The initial model development was based in part on the previous USGS analysis that defined the hydrogeologic-unit surfaces (Smolensky and others, 1990). Initial estimates of aquifer permeability were derived from an analysis of the existing deep lithologic boring data that was used to develop the lithologic texture model described in Walter and Finkelstein (2020).
Aquifer recharge:
Recharge is the sole source of water to the Long Island aquifer system and it consists of four main components:
- Natural recharge from precipitation
- Redirected recharge from impervious surfaces
- Returnflow from domestic-septic systems
- Returnflow from leaky infrastructure
Groundwater withdrawals:
Approximately 460 million gallons per day of water is pumped from the Long Island aquifer system, and this pumping consists of four main components:
- Public supply
- Remediation pumping from contaminated sites
- Farm irrigation
- Industrial and commercial facilities.
Public-supply withdrawals account for nearly (94%) of the groundwater pumped from the aquifer system for current (2015) conditions (Walter and others, 2024).
Boundary between fresh and saline groundwater:
The boundary between fresh and saline groundwater – otherwise referred to as the freshwater/saltwater interface, represents the lateral extent of the fresh-groundwater system in the first phase of model development. In this first phase, the freshwater/saltwater interface assigned in the steady-state, freshwater-only model (Walter and others, 2020) model was based on the current information obtained from previous sampling and geophysical logging (Stumm and others, 2020). This information was obtained as part of this ongoing investigation (see “Saltwater-Interface Mapping”). Subsequent updates to the groundwater flow included a numerical solver (MODFLOW 6 Transport) that allowed for the simulation of the position and movement of the freshwater/saltwater interface in response to changes in hydrologic stresses over time (Walter and others, 2024).
Model calibration:
Once the groundwater model datasets were assembled, the model simulated water levels and streamflows are compared to observed data to determine how well the model matches real-world conditions. Often times, model parameters such as recharge and hydraulic properties need to be adjusted to improve this match by a process referred to as “model calibration”.
Phase I Model Scenarios:
The calibrated model was used to simulate a set of scenarios to examine the potential effects of changes in pumping, recharge, and sea-level position. These scenarios then were compared to base case simulations that included average-annual and average-seasonal conditions for the 2010–2019 period. These base-case simulations are referred to as scenario_1a and scenario_1aa, respectively.
The remaining scenarios are as follows:
- Scenario_2a: Peak season pumping increased by 15 percent.
- Scenario_2b: Scenario_2a plus natural recharge increased by 10 percent.
- Scenario_3a: Peak season pumping decreased by 15 percent.
- Scenario_3b: Scenario_3a plus natural recharge increased by 10 percent.
- Scenario_4a,b,c: Sea-level position increased by 3, 6, and 9 feet.
- Scenario_5: Pumping increased by 20 percent and natural recharge reduced by 20 percent in order to represent a 5-year drought.
- Scenario_6: Natural recharge increased by 10 percent.
- Scenario_7: Reactivation of the Jamaica Water Company production wells at 62 Mgal/d for 10 years.
The simulation results from these scenarios can be viewed graphically in the interactive model results viewer below:
Model Results Viewer
Numerical models provide a means to synthesize existing hydrogeologic information into an internally consistent mathematical representation of a real system or process, and thus are useful tools for testing and improving conceptual models or hypotheses of groundwater flow systems. The goal of this effort is to develop a regional model for the Long Island aquifer system to simulate changes in water levels, streamflows, the position and movement of the boundary between fresh and saline groundwater, assess groundwater age distributions, and to determine regional hydrologic response to changes in anthropogenic and natural stresses in order to assist in the determination of the groundwater sustainability of the Long Island aquifer system.
Modeling Approach:
Pertinent hydrogeologic information is needed to develop a conceptual model of how water enters, moves through, and ultimately leaves the aquifer system. The conceptual model then is used as the foundation for the construction of a 3D-numerical model capable of simulating groundwater flow conditions throughout the Long Island aquifer system. The collection of digital-hydrogeologic data in the study area are needed for proper model development, including elevations and extents of hydrostratigraphic units, groundwater salinity distribution, groundwater withdrawals, temperature and precipitation distributions, recharge and impervious-surface distribution, and additional information necessary for accurate model calibration including groundwater levels and streamflow.
The sources (or inflows) and sinks (or outflows) of water to the Long Island aquifer system can be illustrated schematically (fig. 1) to show how water enters, flows through, and exits this aquifer system). The sole source of water for predevelopment conditions is recharge from precipitation. All of this water entering the aquifer system is balanced by water leaving the aquifer system. Water leaves the aquifer system as discharge to streams, discharge to shallow-coastal waters, and deep-subsea discharge farther offshore in the Atlantic Ocean. An additional loss of water from the aquifer system for post- development conditions is water removed by groundwater withdrawals. Depending on the type of water use, and whether (or how much) a community is sewered a substantial amount of the water withdrawn from the aquifer system may be returned in the form of wastewater-return flow through domestic septic systems.
Modeling Components:
The main components associated with the model development include (1) hydrogeologic framework, (2) aquifer recharge, (3) water use, (4) freshwater/saltwater boundary, and (5) model calibration.
Hydrogeologic Framework:
The hydrogeologic framework consists of the extents and thicknesses of the major hydrogeologic units and the hydraulic properties of the sediments that comprise these units. The initial model development was based in part on the previous USGS analysis that defined the hydrogeologic-unit surfaces (Smolensky and others, 1990). Initial estimates of aquifer permeability were derived from an analysis of the existing deep lithologic boring data that was used to develop the lithologic texture model described in Walter and Finkelstein (2020).
Aquifer recharge:
Recharge is the sole source of water to the Long Island aquifer system and it consists of four main components:
- Natural recharge from precipitation
- Redirected recharge from impervious surfaces
- Returnflow from domestic-septic systems
- Returnflow from leaky infrastructure
Groundwater withdrawals:
Approximately 460 million gallons per day of water is pumped from the Long Island aquifer system, and this pumping consists of four main components:
- Public supply
- Remediation pumping from contaminated sites
- Farm irrigation
- Industrial and commercial facilities.
Public-supply withdrawals account for nearly (94%) of the groundwater pumped from the aquifer system for current (2015) conditions (Walter and others, 2024).
Boundary between fresh and saline groundwater:
The boundary between fresh and saline groundwater – otherwise referred to as the freshwater/saltwater interface, represents the lateral extent of the fresh-groundwater system in the first phase of model development. In this first phase, the freshwater/saltwater interface assigned in the steady-state, freshwater-only model (Walter and others, 2020) model was based on the current information obtained from previous sampling and geophysical logging (Stumm and others, 2020). This information was obtained as part of this ongoing investigation (see “Saltwater-Interface Mapping”). Subsequent updates to the groundwater flow included a numerical solver (MODFLOW 6 Transport) that allowed for the simulation of the position and movement of the freshwater/saltwater interface in response to changes in hydrologic stresses over time (Walter and others, 2024).
Model calibration:
Once the groundwater model datasets were assembled, the model simulated water levels and streamflows are compared to observed data to determine how well the model matches real-world conditions. Often times, model parameters such as recharge and hydraulic properties need to be adjusted to improve this match by a process referred to as “model calibration”.
Phase I Model Scenarios:
The calibrated model was used to simulate a set of scenarios to examine the potential effects of changes in pumping, recharge, and sea-level position. These scenarios then were compared to base case simulations that included average-annual and average-seasonal conditions for the 2010–2019 period. These base-case simulations are referred to as scenario_1a and scenario_1aa, respectively.
The remaining scenarios are as follows:
- Scenario_2a: Peak season pumping increased by 15 percent.
- Scenario_2b: Scenario_2a plus natural recharge increased by 10 percent.
- Scenario_3a: Peak season pumping decreased by 15 percent.
- Scenario_3b: Scenario_3a plus natural recharge increased by 10 percent.
- Scenario_4a,b,c: Sea-level position increased by 3, 6, and 9 feet.
- Scenario_5: Pumping increased by 20 percent and natural recharge reduced by 20 percent in order to represent a 5-year drought.
- Scenario_6: Natural recharge increased by 10 percent.
- Scenario_7: Reactivation of the Jamaica Water Company production wells at 62 Mgal/d for 10 years.
The simulation results from these scenarios can be viewed graphically in the interactive model results viewer below: